CN116934837B

CN116934837B - Object detection processing method, device, system, electronic equipment and storage medium

Info

Publication number: CN116934837B
Application number: CN202311175688.2A
Authority: CN
Inventors: 唐志宏; 郑剑杰; 卢春明
Original assignee: Shanghai Chaoqun Testing Technology Co ltd
Current assignee: Shanghai Chaoqun Testing Technology Co ltd
Priority date: 2023-09-13
Filing date: 2023-09-13
Publication date: 2023-11-24
Anticipated expiration: 2043-09-13
Also published as: CN116934837A

Abstract

The application provides an object detection processing method, an object detection processing device, an object detection processing system, electronic equipment and a storage medium, wherein the method comprises the following steps: receiving visible light detection data of an object to be detected sent by a visible light detection device, wherein the visible light detection data comprises at least one of the following components: shape information, actual weight, color; determining the target category of the object to be detected according to the visible light detection data; receiving the multi-energy detection data of the object to be detected, which is sent by a multi-energy X-ray detection device; according to the target category, acquiring a plurality of sub-category energy attenuation models corresponding to the target category; and determining a sub-category to be verified of the object to be tested according to the multi-energy detection data and the sub-category energy attenuation models, and determining a target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data of the object to be tested. The comprehensive detection of the object can be realized, the loss of object information is avoided, and the detection efficiency is improved.

Description

Object detection processing method, device, system, electronic equipment and storage medium

Technical Field

The present application relates to the field of X-ray detection technology, and in particular, to an object detection processing method, device, system, electronic device, and storage medium.

Background

The X-ray imaging technique acquires an image of an object by using X-rays. X-ray imaging captures an image of the inside of an object to be inspected using a non-invasive method, thereby irradiating X-rays onto the object to be inspected, and detecting the X-rays transmitted by the object to be inspected. Therefore, the X-ray imaging technology is widely applied to the field of physical research and is a main mode of material diagnosis.

In the prior art, a dual-energy X-ray imaging apparatus may be used to detect an object to be examined. The dual-energy X-ray has only high energy and low energy, and for objects with large density, the high-energy X-ray cannot penetrate, substances with lower density directly penetrate, so that the dual-energy X-ray is relatively single, only the detected objects in a certain range can be detected, the detected objects are limited, and the condition that detection information is lost possibly exists. Meanwhile, manual participation in analysis processing is needed in the detection process by using the dual-energy X-ray imaging equipment, so that the detection efficiency is low.

Disclosure of Invention

The application aims to overcome the defects in the prior art and provide an object detection processing method, device and system, which improve the accuracy and efficiency of object detection.

In order to achieve the above purpose, the technical scheme adopted by the embodiment of the application is as follows:

in a first aspect, an embodiment of the present application provides an object detection processing method, where the method includes:

receiving visible light detection data of an object to be detected sent by a visible light detection device, wherein the visible light detection data comprises at least one of the following components: shape information, actual weight, color;

determining the target category of the object to be detected according to the visible light detection data;

receiving the multi-energy detection data of the object to be detected sent by a multi-energy X-ray detection device, wherein the multi-energy detection data comprises: the input energy, the output energy and the thickness of the object to be measured under various energies and various angles;

according to the target category, acquiring a plurality of sub-category energy attenuation models corresponding to the target category, wherein each sub-category energy attenuation model is used for representing the conversion condition of the output energy of each sub-category object under the target category along with the thickness;

and determining a sub-category to be verified of the object to be tested according to the multi-energy detection data and the sub-category energy attenuation models, and determining a target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data of the object to be tested.

Optionally, the determining, according to the visible light detection data, the target class of the object to be detected includes:

determining the estimated volume of the object to be detected according to the shape information of the object to be detected;

determining the estimated density of the object to be detected according to the estimated volume of the object to be detected and the actual weight;

inputting the shape information, the estimated volume and the estimated density of the object to be detected into a class identification model constructed in advance to obtain the target class of the object to be detected.

Optionally, the determining the sub-category to be verified of the object to be tested according to the multi-energy detection data and the multiple sub-category energy attenuation models includes:

constructing an actual energy attenuation curve of the object to be measured under each X-ray energy according to the input energy, the output energy and the thickness of the object to be measured under various angles under each X-ray energy;

and determining the sub-category to be verified of the object to be tested according to the actual energy attenuation curves of the object to be tested under each X-ray energy and the sub-category energy attenuation models, wherein each sub-category energy attenuation model comprises standard energy attenuation curves under various X-ray energies.

Optionally, the constructing an actual energy attenuation curve of the object to be measured under each X-ray energy according to the input energy, the output energy and the thickness of the object to be measured under various angles under each X-ray energy includes:

determining an attenuation coefficient of the object to be measured under the current X-ray energy according to the input energy, the output energy and the thickness of the object to be measured under various angles of the current X-ray energy;

and constructing an actual energy attenuation curve of the object to be measured under each X-ray energy according to the output energy and thickness of the object to be measured under various angles under the current X-ray energy and the attenuation coefficient under the current X-ray energy.

Optionally, the determining, according to the input energy, the output energy and the thickness of the object to be measured under the current X-ray energy at various angles, the attenuation coefficient corresponding to each thickness of the object to be measured under the current X-ray energy includes:

and inputting the input energy, the output energy and the thickness into a preset energy attenuation formula to obtain an attenuation coefficient corresponding to the thickness.

Optionally, the determining the sub-category to be verified of the object to be tested according to the actual energy attenuation curve of the object to be tested under each X-ray energy and the energy attenuation models of the sub-categories includes:

Traversing the actual energy attenuation curve under each X-ray energy, aiming at the traversed current actual energy attenuation curve under the current X-ray energy:

fitting and matching the current actual energy attenuation curve with the standard energy attenuation curve under the current X-ray energy in each sub-category energy attenuation model to obtain matching degree information of the current actual energy attenuation curve and the standard energy attenuation curve under the current X-ray energy in each sub-category energy attenuation model;

determining a matched sub-category under the current X-ray energy according to the matching degree information of the current actual energy attenuation curve and the standard energy attenuation curve under the current X-ray energy in each sub-category energy attenuation model;

after the actual energy attenuation curve under each X-ray energy is traversed, determining the sub-category to be verified of the object to be tested according to the matched sub-categories under all the X-ray energy.

Optionally, the determining the target subcategory of the object to be tested according to the subcategory to be verified includes:

determining the density to be verified and the volume to be verified according to the subcategory to be verified;

determining the weight to be verified of the object to be tested according to the density to be verified and the volume to be verified;

And determining the target subcategory of the object to be tested according to the weight to be verified and the actual weight.

Optionally, the determining the target subcategory of the object to be tested according to the weight to be verified and the actual weight includes:

if the difference value between the weight to be verified and the actual weight is smaller than a preset threshold value, the sub-category to be verified is used as the target sub-category;

and if the difference value between the weight to be verified and the actual weight is greater than or equal to the preset threshold value, sending a re-detection instruction to the visible light detection device and the X-ray detection device, receiving new visible light detection data sent by the visible light detection device after re-detection and new multi-energy detection data sent by the X-ray detection device after re-detection, and re-determining the target subcategory of the object to be detected according to the new visible light detection data and the new multi-energy detection data.

In a second aspect, an embodiment of the present application further provides an object detection processing apparatus, including:

the receiving module is used for receiving the visible light detection data of the object to be detected, which is sent by the visible light detection device, wherein the visible light detection data comprises at least one of the following components: shape information, actual weight, color;

The determining module is used for determining the target category of the object to be detected according to the visible light detection data;

the receiving module is used for receiving the multi-energy detection data of the object to be detected, which is sent by the multi-energy X-ray detection device, and the multi-energy detection data comprises: the input energy, the output energy and the thickness of the object to be measured under various energies and various angles;

the acquisition module is used for acquiring a plurality of sub-category energy attenuation models corresponding to the target category according to the target category, wherein each sub-category energy attenuation model is used for representing the conversion condition of the output energy of each sub-category object under the target category along with the thickness;

the determining module is used for determining the sub-category to be verified of the object to be tested according to the multi-energy detection data and the sub-category energy attenuation models, and determining the target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data of the object to be tested.

Optionally, the determining module is specifically configured to:

In a third aspect, an embodiment of the present application further provides an object detection processing system, including: a visible light detection device, a multi-energy X-ray detection device, a classification device and a cloud server;

The visible light detection device, the multi-energy X-ray detection device and the classification device are arranged in a scene where an object to be detected is located, and the object to be detected sequentially passes through the visible light detection device, the multi-energy X-ray detection device and the classification device;

the cloud server is in communication connection with the visible light detection device, the multi-energy X-ray detection device and the classification device, and detects the object machine to be detected based on the method in the first aspect.

In a fourth aspect, an embodiment of the present application further provides an electronic device, including: the object detection processing method comprises a processor, a storage medium and a bus, wherein the storage medium stores program instructions executable by the processor, when an application program runs, the processor and the storage medium are communicated through the bus, and the processor executes the program instructions to execute the steps of the object detection processing method in the first aspect.

In a fifth aspect, an embodiment of the present application further provides a computer readable storage medium, on which a computer program is stored, the computer program being read and executed the steps of the object detection processing method described in the first aspect.

The beneficial effects of the application are as follows:

according to the object detection processing method, the device, the system, the electronic equipment and the storage medium, the multi-energy multi-angle scanning is carried out on the object to be detected to obtain the multi-energy detection data of the object to be detected, comprehensive detection of complex objects with various densities can be achieved, the loss of object information is avoided, the multi-energy detection data obtained through detection and the multi-subcategory energy attenuation models under the target category are utilized to determine the subcategory to be verified of the object to be detected, and the target subcategory of the object to be detected is determined according to the subcategory to be verified and the visible light detection data, so that manual participation analysis in the detection process can be avoided, the detection efficiency is improved, and the large-scale production line detection is achieved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of an object detection processing system according to an embodiment of the present application;

fig. 2 is a schematic flow chart of an object detection processing method according to an embodiment of the present application;

FIG. 3 is a schematic diagram of the output energy as a function of thickness according to an embodiment of the present application;

FIG. 4 is a schematic diagram of an energy attenuation model according to an embodiment of the present application;

FIG. 5 is a schematic flow chart of determining a target class according to an embodiment of the present application;

FIG. 6 is a schematic diagram of a method for creating a metal data model according to an embodiment of the present application;

FIG. 7 is a schematic diagram of a database for creating a metal model according to an embodiment of the present application;

FIG. 8 is a schematic diagram of a class identification model according to an embodiment of the present application;

FIG. 9 is a flowchart of a method for determining a sub-category to be verified of an object to be tested according to an embodiment of the present application;

FIG. 10 is a schematic flow chart of an energy attenuation curve construction according to an embodiment of the present application;

FIG. 11 is a schematic flow chart of determining a sub-category to be verified according to an embodiment of the present application;

FIG. 12 is a schematic flow chart of determining a target sub-category according to an embodiment of the present application;

FIG. 13 is a schematic diagram of an apparatus for detecting an object according to an embodiment of the present application;

Fig. 14 is a block diagram of an electronic device according to an embodiment of the present application.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for the purpose of illustration and description only and are not intended to limit the scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this disclosure, illustrates operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to or removed from the flow diagrams by those skilled in the art under the direction of the present disclosure.

In addition, the described embodiments are only some, but not all, embodiments of the application. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present application.

It should be noted that the term "comprising" will be used in embodiments of the application to indicate the presence of the features stated hereafter, but not to exclude the addition of other features.

Fig. 1 is a schematic structural diagram of an object detection processing system according to an embodiment of the present application, where the system includes a visible light detection device, a multi-energy X-ray detection device, a classification device, and a cloud server, where the visible light detection device and a conveyor belt in an area where the visible light detection device is located together form a preprocessing system, the multi-energy X-ray detection device and the conveyor belt in an area where the multi-energy X-ray detection device is located together form a detection system, and the classification device and the conveyor belt in an area where the classification device is located together form a post-processing system.

Optionally, when the object to be detected needs to be detected, the object to be detected may be placed on a transmission belt on a predetermined track in fig. 1, and when the object to be detected is transmitted to a range detectable by the visible light detection device through the transmission belt, the visible light detection device performs visible light scanning on the object to be detected to obtain visible light detection data of the object to be detected, the obtained visible light detection data is sent to the cloud server, and the cloud server performs preprocessing analysis according to the received visible light detection data; when the visible light detection device finishes scanning an object to be detected, the object to be detected is continuously transmitted to the multi-energy X-ray detection device, and when the object to be detected is transmitted to the multi-energy X-ray detection device within the detectable range, the multi-energy X-ray detection device detects the object to be detected to obtain multi-energy detection data, the detected multi-energy detection data are sent to the cloud server, the cloud server analyzes the received multi-energy detection data, and the visible light detection data and the analysis result of the multi-energy detection data are combined to obtain the target subcategory of the object to be detected. Further, the cloud server can transmit the object to be measured to the corresponding sub-packaging table through interaction with the sorting device. For example, fig. 1 may include 2 sub-loading stations, which may be a sub-loading station of category 1 and a sub-loading station of category 2, and if the target sub-category of the object to be measured is category 1, when the object to be measured is transferred to the area where the post-processing system is located, the classifying device transfers the object to be measured to the sub-loading station of category 1 in a transmission manner, so as to classify the object to be measured. Alternatively, the sorting device may clamp the objects to be tested to the sub-packaging table by using a robot arm.

Fig. 2 is a flow chart of an object detection processing method according to an embodiment of the present application, as shown in fig. 2, an execution body of the method is a cloud server as described above, and the method may include:

s101, receiving visible light detection data of an object to be detected, which is sent by a visible light detection device.

Wherein, the visible light detection data can comprise at least one of the following: shape information, actual weight, color, etc.

Optionally, when the object to be detected is within the detection range of the visible light detection device in fig. 1, the visible light detection device may scan the object to be detected by using the visible light imaging device, and send the visible light detection data obtained by scanning to the cloud server.

Optionally, the visible light imaging device may perform multi-angle scanning on the object to be measured, specifically, the multi-angle visible light imaging device may be used for one-time imaging, and may include one-time imaging at multiple angles such as vertical, horizontal, and four vertex angles oblique; the visible light imaging device can be used for circumferential scanning, and the two dimensions of horizontal scanning and vertical scanning can be included; and obtaining a visible light image of the object to be detected, wherein the visible light image can comprise shape information, color information and the like of the object to be detected.

Optionally, the object to be measured can be weighed by the weighing device in the visible light detection device, so as to obtain the actual weight of the object to be measured.

S102, determining the target category of the object to be detected according to the visible light detection data.

Optionally, after the cloud server receives the visible light data of the object to be detected sent by the visible light detection device, a preset method may be used to determine a target class of the object to be detected according to the visible light detection data.

The target class may be, for example, a large class of metal, ore, wood, or the like. Based on the target class, the target subcategory of the object to be measured can be further determined based on the following steps. For example, when the target class is metal, the target subcategory may be iron, copper, aluminum, or the like.

S103, receiving the multi-energy detection data of the object to be detected, which is sent by the multi-energy X-ray detection device.

Wherein, the multi-energy detection data may include: the input energy, the output energy and the thickness of the object to be measured under various energies and various angles.

Exemplary, for example, input energy, output energy, and thickness of an object under test at a horizontal angle of kV1 energy; input energy, output energy, and thickness at a vertical angle at kV1 energy; input energy, output energy and thickness at oblique angle at kV1 energy.

Optionally, when the object to be detected is transferred to the detectable range of the multi-energy X-ray detecting device, the multi-energy X-ray imaging device may be used to scan the object to be detected, where the multi-energy X-ray imaging device may include a multi-energy X-ray generator and a multi-energy X-ray detector, and the multi-energy X-rays generated by the multi-energy X-ray generator are irradiated to the object to be detected, and when the multi-energy X-rays penetrate through the object to be detected, the multi-energy X-ray detector detects energy of the multi-energy X-rays penetrating through the object to be detected.

The input energy may refer to energy of an X-ray source generated by an X-ray generator, the output energy refers to energy of X-rays transmitted through an object to be measured, and the thickness may refer to thickness of the object through which the X-rays pass under different angles.

Optionally, the multi-energy X-ray generator can generate rays with multiple energies, and the voltage and current of the X-rays generated by the multi-energy X-ray generator can be set according to actual requirements, for example, the voltage can be set to any voltage between 30kV and 130kV, such as 30kV, 50kV, 70kV, 90kV, 110kV, 130kV and the like.

Optionally, the object to be detected can be scanned at multiple angles under different energies, and when the object to be detected is scanned at multiple angles, a multi-energy X-ray imaging device can be utilized to scan the object to be detected at multiple angles under the support of a motion mechanism, for example, the multi-angle scanning can include but not limited to horizontal, vertical, oblique and other angles; the multi-energy X-ray imaging equipment can be used for respectively scanning horizontal, vertical, oblique incidence and other angles, and the multi-energy detection data of the object to be detected obtained by scanning of each multi-energy X-ray imaging equipment are sent to the cloud server; specifically, the multiple multi-energy X-ray imaging devices can be respectively arranged at different positions such as a horizontal position, a vertical position, an oblique position and the like, and under one energy, the multiple multi-energy X-ray imaging devices respectively scan an object to be measured at different angles.

Optionally, the object to be measured may be made of a plurality of different material materials, where each material includes a density of the material, and then the object to be measured may include a plurality of material distributions and a plurality of density distributions, and by performing multi-energy and multi-angle scanning on the object to be measured, a situation that output energy of reference materials with different thicknesses under different energy X-rays is changed along with the thickness can be obtained, where the reference materials refer to each material that constitutes the object to be measured. When the X-rays with the same energy are used for scanning, the situation that the X-rays directly penetrate or cannot penetrate can possibly occur, and when the X-rays directly penetrate, the X-rays penetrate through the substances and the energy loss is small; when the X-rays cannot penetrate, the X-rays are so small that the X-ray detector detects the X-rays after passing through the substance that the X-ray energy passing through the substance cannot be detected. Therefore, the multi-energy and multi-dimensional scanning is utilized to obtain the multi-energy detection data of the object to be detected under the X-rays with different energies, so that different material information and distribution in the object to be detected can be reflected, and the defect of the information of the object to be detected is avoided.

S104, acquiring a plurality of sub-category energy attenuation models corresponding to the target category according to the target category.

The energy attenuation model of each sub-category can refer to the transformation condition of the output energy of each sub-category of the object under the target category along with the thickness.

Alternatively, for the same substance of different thickness, under one energy X-ray, the output energy changes regularly with the change of thickness. For a certain gold, fig. 3 may be used to represent the relationship between the thickness and the output energy, fig. 3 is a schematic diagram of the conversion of the output energy with the thickness according to the embodiment of the present application, and the data in fig. 3 is curve-fitted to obtain the attenuation curve in fig. 4, and fig. 4 is a schematic diagram of an energy attenuation model according to the embodiment of the present application.

Alternatively, a plurality of sub-category energy attenuation models may be included under each target category, and for example, for metal categories, an iron energy attenuation model, a copper energy attenuation model, a silver energy attenuation model, a manganese energy attenuation model, a gold energy attenuation model, etc. may be included; for the ore class, for example, pyrite energy decay models, chalcopyrite energy decay models, galena energy decay models, and the like may be included.

S105, determining a sub-category to be verified of the object to be tested according to the multi-energy detection data and the energy attenuation models of the sub-categories, and determining a target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data.

Optionally, the sub-category to be verified of the object to be tested may be determined according to the multi-energy detection data of the object to be tested obtained in S103 and the multiple sub-category energy attenuation models under the target category of the object to be tested obtained in S104, and the target sub-category of the object to be tested may be determined according to the sub-category to be verified. The sub-category to be verified and the target sub-category may refer to, for example, that the object to be tested belongs to sub-categories such as iron, copper, aluminum, etc. under the metal category.

In this embodiment, the multi-energy multi-angle scanning is performed on the object to be detected to obtain multi-energy detection data of the object to be detected, so that comprehensive detection on complex objects with various densities can be realized, the loss of object information is avoided, the multi-energy detection data obtained by detection and the multi-subcategory energy attenuation model under the target category are utilized to determine the subcategory to be verified of the object to be detected, and the target subcategory of the object to be detected is determined according to the subcategory to be verified and the visible light detection data, so that manual participation analysis in the detection process can be avoided, the detection efficiency is improved, and the large-scale production line detection is realized.

Fig. 5 is a schematic flow chart of determining a target class according to an embodiment of the present application, as shown in fig. 5, where determining the target class of the object to be detected according to the visible light detection data in S102 may include:

s201, determining the estimated volume of the object to be detected according to the shape information of the object to be detected.

The shape information may refer to, for example, the size of the object to be measured, such as information of the length, width, height, etc. of the object to be measured. The method can scan the object to be measured at multiple angles to obtain the shape information of the object to be measured at multiple angles, and can determine and obtain the estimated volume of the object to be measured according to the shape information of the object at multiple angles.

S202, determining the estimated density of the object to be detected according to the estimated volume and the actual weight of the object to be detected.

Specifically, a calculation formula in which the product of the density and the volume is equal to the weight may be used to calculate the estimated density of the object to be measured.

S203, inputting the shape information, the estimated volume and the estimated density of the object to be detected into a class identification model which is constructed in advance, so as to obtain the target class of the object to be detected.

The model database of the object of various types can be included in the category identification model, and the shape information, the volume information, the density information and the like of different substances in the category can be included in each category model database. For example, a metal model database, an ore model database, a wood model database and the like can be included in the category identification model, and a metal data model 1, a metal data model 2 and a metal data model 3 can be included in the metal model database; each metal data model is obtained through visible light data fusion training of the metal under different volumes and different shapes.

Optionally, the category identification model may identify the category of the object to be detected according to the input shape information, the estimated volume and the estimated density of the object to be detected. Specifically, shape information, estimated volume and estimated density of the object to be measured can be matched with shape information, volume information and density information of each substance in each model database in the category identification model, and categories of the model databases matched with the shape information, the volume information and the density information of the object to be measured are used as target categories of the object to be measured.

For the object a to be detected, the visible light data of the object a to be detected is respectively matched with the shape information, the volume information and the density information of various metals in a metal model database in the category identification model; respectively matching with shape information, volume information and density information of various stones in a stone model database; respectively matching with shape information, volume information and density information of various woods in a wood model database; if the object A to be measured is matched with the metal 1, the category of the metal model database where the metal 1 is located is taken as the target category of the object to be measured, that is to say, the object to be measured is a metal category.

Optionally, the pre-constructed class identification model may perform multi-angle scanning of visible light on different substances in different classes by using a visible light imaging device in advance to obtain shape information, volume information, density information and the like of each substance, perform fusion training on the scanned information to generate a data model of the substance, and perform fusion training on the data models of different substances in the same class to obtain a class model database.

For example, for metal types, visible light scanning can be performed on the metal 1 in advance, specifically, metal 1 samples with different volumes and different shapes can be selected for scanning, and a metal data model 1 is established; carrying out visible light scanning on the metal 2, and establishing a metal data model 2; and (3) carrying out visible light scanning on the metal n, establishing a metal data model n, and then carrying out fusion training on the metal data model 1, the metal data model 2 and the metal data model n to finally obtain a metal model database. The metal model database can be derived in particular from fig. 6 and 7. FIG. 6 is a schematic diagram of a method for creating a metal data model according to an embodiment of the present application; fig. 7 is a schematic diagram of a database for creating a metal model according to an embodiment of the present application.

The class identification model may be established through the process shown in fig. 8, and fig. 8 is a schematic diagram of a class identification model according to an embodiment of the present application. Similarly, for other types of visible light data models, such as an ore visible light data model, a wood visible light data model, etc., the corresponding visible light data model can be built according to the building method of the metal visible light data model, which is not described herein. And then, obtaining a category identification model through fusion training of the visible light data models of different categories.

In this embodiment, the class recognition model is used to determine the target class of the object to be detected according to the visible light detection data of the object to be detected, so that manual participation in analysis can be avoided, the detection efficiency of the object to be detected is improved, and meanwhile, the energy attenuation models of multiple sub-classes under the target class can be more conveniently called in the subsequent detection system.

Fig. 9 is a flowchart of a method for determining a sub-category to be verified of an object to be tested according to an embodiment of the present application, as shown in fig. 9, the determining the sub-category to be verified of the object to be tested according to the multi-energy detection data and the multiple sub-category energy attenuation models in S105 may include:

S301, constructing an actual energy attenuation curve of the object to be measured under each X-ray energy according to the input energy, the output energy and the thickness of the object to be measured under various angles under each X-ray energy.

The actual energy attenuation curve refers to the transformation condition of the output energy of the object along with the thickness. The attenuation of the output energy is also different for the same thickness of material at different X-ray energies, and the attenuation of the output energy is changed regularly for the same material of different thickness at the same X-ray energy, wherein the material refers to a single material of material.

Optionally, the object to be measured is scanned at multiple angles under the energy of an X-ray, so that different thicknesses of the object to be measured under different angles can be obtained. Therefore, the conversion condition of the output energy of the object to be measured under each X-ray energy along with the thickness can be constructed.

S302, determining the sub-category to be verified of the object to be tested according to the actual energy attenuation curve of the object to be tested under each X-ray energy and the energy attenuation models of the sub-categories.

Optionally, as can be seen from the foregoing, each sub-category energy attenuation model may refer to a transformation condition of output energy of each sub-category object under the target category with thickness, and then an actual energy attenuation curve of the object to be tested under each type of X-ray energy may be fitted with a plurality of sub-category energy attenuation models, so that a sub-category of the sub-category energy attenuation model with the highest fitting degree may be used as a sub-category to be verified of the object to be tested.

In this embodiment, the transformation condition of the output energy of the object to be tested along with the thickness is obtained by performing multi-energy multi-angle scanning on the object to be tested, so that the sub-class to be verified of the object to be tested is obtained by fitting according to the multiple sub-class energy models, and the manual participation in analysis can be avoided, and the detection efficiency and the detection accuracy are improved.

Fig. 10 is a schematic flow chart of an energy attenuation curve construction provided in an embodiment of the present application, as shown in fig. 10, in the step S301, according to input energy, output energy and thickness of an object to be measured under various angles under each X-ray energy, constructing an actual energy attenuation curve of the object to be measured under each X-ray energy may include:

s401, determining an attenuation coefficient of the object to be measured under the current X-ray energy according to the input energy, the output energy and the thickness of the object to be measured under various angles under the current X-ray energy.

Wherein the current X-ray energy refers to any one of a plurality of X-ray energies.

Specifically, if the object to be measured is an elemental object, for example, a cuboid elemental metal element, multiple X-ray energy multi-angle scans can be performed on the object to be measured, so as to obtain input energy, output energy and thickness under each angle. The object to be measured can be scanned under the angle x, and the multi-energy data information of the object to be measured under the angle x can be obtained by scanning the angle x and the angle y Output energy->The value of the thickness d of the X-ray passing through the object under the angle X can be obtained by scanning under the angle y, and the attenuation coefficient of the object to be measured under the current X-ray energy can be obtained.

Optionally, the input energy, the output energy and the thickness of the single-substance object at other angles under the current X-ray energy can be scanned according to the scanning method of the X-angle, so that the thickness of the object to be measured at different angles under the current X-ray energy can be obtained. Because the attenuation coefficients of the same object to be measured with different thicknesses under the same X-ray energy are the same, the attenuation coefficient of the object to be measured under the current X-ray energy can be calculated according to the input energy, the output energy and the thickness of any angle of the object to be measured under the current X-ray energy.

S402, constructing an actual energy attenuation curve of the object to be measured under each X-ray energy according to the output energy and thickness of the object to be measured under various angles under the current X-ray energy and the attenuation coefficient under the current X-ray energy.

Alternatively, for the above elemental objects, a thickness of a plurality of angles may be obtained、/>、/>Because the attenuation coefficients of the same substance with different thicknesses under the same energy are the same, and the energy attenuation changes regularly, the actual energy attenuation curve under the current X-ray energy can be constructed according to the output energy, the thickness and the attenuation coefficient under the current X-ray energy under different angles. The actual energy attenuation curve can also be represented by the attenuation curve diagram shown in fig. 4, wherein the object to be measured in fig. 4 is a single object, and can be obtained in different ways Under the X-ray energy, the actual energy attenuation curve refers to the attenuation coefficient +.>。

In this embodiment, by constructing an actual energy attenuation curve of the object to be detected under each type of X-ray energy, the information of the object to be detected obtained by monitoring can be more comprehensive, and the loss of information is avoided.

Optionally, in S401, determining the attenuation coefficient corresponding to each thickness of the object to be measured under the current X-ray energy according to the input energy, the output energy and the thickness of the object to be measured under the current X-ray energy under the multiple angles may include:

optionally, the input energy, the output energy and the thickness are input into a preset energy attenuation formula, and an attenuation coefficient corresponding to the thickness is obtained.

Specifically, taking the simple substance object in S401 as an example, the input energy of the X-ray obtained at the X-angle isThe output energy is +.>And X-rays pass through the thickness of the object is +.>Input energy is +.>The output energy isThickness ∈of object>Carry-in energy decay formula->In (1)To obtain attenuation coefficient corresponding to X-ray passing through the thickness of the object under the angle X, wherein the input energy is +. >The output energy is +.>Thickness->Are all known values, can be obtained by scanning the object to be measured, and can be calculated to obtain the attenuation coefficient +.>Is a numerical value of (2). Wherein the attenuation coefficient obtained by scanning the single object at other angles is the same as the attenuation coefficient of the x angle, and can be increased by the attenuation coefficient +.>The name of the elemental object and thus the density of the elemental object can be known.

For example, for an object to be tested with two kinds of elemental materials overlapped, for example, under an X angle, multiple X-ray energies can be used for scanning under the X angle to obtain multi-energy detection data of the object to be tested, for example, three kinds of X-ray energies of kV1, kV2 and kV3 are used for scanning the object to be tested, and the following energy attenuation formulas of the object to be tested under the three kinds of energies can be obtained:

(I)

(II)

(III)

the unknowns are as follows for (I), (II) and (III)、/>、/>、/>、/>、、/>、/>Wherein->、/>And X-rays obtained by y-angle scanning pass through the object to be measured>(by counting pixels, considered known) constitutes a new equation:

(IV)

wherein,for X-rays to pass through the thickness of the elementary object 1 in the object to be measured, +.>If X-rays penetrate through the thickness of the simple substance object 2 in the object to be measured, substituting the formula (IV) into the formulas (I) - (III) In (2) can be given the energy of kV1 +.>、/>+.>、/>And +.about.3 under kV3 energy>、/>. Wherein (1)>、/>、/>Refers to the attenuation coefficient of the elemental object 1 at different energies, +.>、/>、The attenuation coefficients of the single-substance object 2 under different energies are referred to, and it is worth to say that the attenuation coefficients of the single-substance object with the same thickness under different X-ray energies are different, that is, the attenuation coefficients of the same single-substance object under different energies are different, but all correspond to the same single-substance object.

For example, the object to be measured with two kinds of simple substance materials overlapped is taken as an example, and the same is repeated at the angle x, namely kV1 and kV2The y angle of the three X-ray energies of kV3 can be obtained、/>、/>、/>、、/>、/>、/>The method comprises the steps of carrying out a first treatment on the surface of the At z angle +.>、/>、/>、/>、/>、、/>、/>. The complex object to be detected is scanned through multiple energies and multiple angles, so that the information of the scanned object is more comprehensive.

Fig. 11 is a schematic flow chart of determining a sub-category to be verified according to an embodiment of the present application, as shown in fig. 11, in the step S302, determining the sub-category to be verified of the object to be tested according to an actual energy attenuation curve of the object to be tested under each type of X-ray energy and a plurality of sub-category energy attenuation models may include:

S501, traversing the actual energy attenuation curve under each X-ray energy, and aiming at the current actual energy attenuation curve under the traversed current X-ray energy: fitting the current actual energy attenuation curve with the standard energy attenuation curve under the current X-ray energy in the energy attenuation model of each sub-category to obtain the matching degree information of the current actual energy attenuation curve and the standard energy attenuation curve under the current X-ray energy in the energy attenuation model of each sub-category.

From the above, the energy attenuation model of each sub-category refers to the transformation of the output energy of each sub-category of the object with the thickness under the target category. By scanning the object to be measured with different X-ray energies, the actual energy attenuation curve of the object to be measured at each X-ray energy can be obtained.

For example, if the object to be measured is a metal type, the metal type includes a metal 1 energy attenuation model, a metal 2 energy attenuation model, a metal 3 energy attenuation model, and a metal n energy attenuation model. The metal energy attenuation models of the different sub-categories are shown in fig. 4, and include the conversion condition of the output energy of the object under different energies with the thickness, that is, the standard energy attenuation curve of the metal of each sub-category under different energies.

For example, if the current X-ray energy of the traversed object to be measured is kV1, fitting an actual energy attenuation curve under kV1 energy with a standard energy attenuation curve under kV1 energy in each sub-category energy attenuation model to obtain matching degree information of the actual energy attenuation curve under kV1 energy and the standard energy attenuation curve under kV1 energy in each sub-category energy attenuation model.

Optionally, as known from S402, the actual energy attenuation curve of the object under test under the current X-ray energy is constructed by the thicknesses of different angles under the current X-ray energy, the output energy and the attenuation coefficient under the current X-ray energy, so that the thickness of the X-angle can be utilized under kV1 energy for the X-angleAnd attenuation coefficient->Fitting the thickness and attenuation coefficient in a standard energy attenuation curve under kV1 energy in each sub-category energy attenuation model to obtain +.>And attenuation coefficient->Matching degree information of thickness and attenuation coefficient in a standard energy attenuation curve under kV1 energy in each sub-category energy attenuation model; thickness +.x-angle can be utilized at kV2 energy>And attenuation coefficient->Fitting the thickness and attenuation coefficient in a standard energy attenuation curve under kV2 energy in each sub-category energy attenuation model to obtain +. >And attenuation coefficient->Matching degree information of thickness and attenuation coefficient in a standard energy attenuation curve under kV1 energy in each sub-category energy attenuation model; thickness +.x-angle can be utilized at kV3 energy>And attenuation coefficient->Fitting the thickness and attenuation coefficient in a standard energy attenuation curve under kV3 energy in each sub-category energy attenuation model to obtain +.>And attenuation coefficient->And matching degree information of thickness and attenuation coefficient in a standard energy attenuation curve under kV3 energy in each sub-category energy attenuation model.

For thickness at y-angleAnd attenuation coefficient->Thickness at z-angle->And attenuation coefficient->And the matching degree information of different thicknesses and attenuation coefficients under the current X-ray energy and the thicknesses and attenuation coefficients in the standard energy attenuation curve under the current X-ray energy in the energy attenuation models of all the sub-categories can be obtained as in the fitting method under the angle X.

S502, determining a matched sub-category under the current X-ray energy according to the matching degree information of the current actual energy attenuation curve and the standard energy attenuation curve under the current X-ray energy in the energy attenuation model of each sub-category.

Optionally, if the matching degree between the current actual energy attenuation curve and the standard energy attenuation curve under the current X-ray energy in the energy attenuation model of each sub-category is highest, the sub-category with the highest matching degree with the current actual energy attenuation curve is used as the matching sub-category of the object to be detected under the current X-ray energy.

For example, if the actual energy decay curve at kV1 energy matches the standard energy decay curve of kV1 for metal 1 most, then metal 1 is considered as the class of matching sub-at kV 1.

For example, continuing with the above elemental object as an example, for an x angle, at a thickness of kV1 energyAnd attenuation coefficientCan obtain optimal +.>And attenuation coefficient->The method comprises the steps of carrying out a first treatment on the surface of the Thickness at kV2 energy +.>And attenuation coefficient->Can obtain optimal +.>And attenuation coefficient->The method comprises the steps of carrying out a first treatment on the surface of the Thickness +.>And attenuation coefficient->Can obtain optimal +.>And attenuation coefficient->Can be represented by the following equations (V) - (VII), wherein equation (VIII)) Fitting the constraint of equations (V) - (VII) to obtain +.>Thickness +.>The same applies.

（Ⅴ）

（Ⅵ）

（Ⅶ）

（Ⅷ）

The fitting method under different X-ray energy under the y-angle and the z-angle is the same as the fitting method under the X-angle, and can be fitted to the attenuation coefficient under the current X-ray energy and the same as the X-angle, for example, under the kV1 energyAnd->，/>And->The method comprises the steps of carrying out a first treatment on the surface of the At kV2 energy, can be fitted to +.>And->，/>And->The method comprises the steps of carrying out a first treatment on the surface of the At kV3 energy, can be fitted to +.>And->，/>And->。

S503, after the actual energy attenuation curve under each X-ray energy is traversed, determining the sub-category to be verified of the object to be tested according to the matched sub-categories under all the X-ray energy.

Specifically, after the actual energy attenuation curve under each X-ray energy is traversed, the sub-category with the largest number of matching sub-categories under all X-ray energy can be used as the sub-category to be verified of the object to be tested.

For example, if the matching sub-category under kV1 energy is metal 1, the matching sub-category under kV2 energy is metal 3, and the matching sub-category under kV3 energy is metal 1, then metal 1 may be used as the sub-category to be verified of the object to be tested.

For example, continuing to take the above elemental objects as examples, if、/>、/>In the standard energy decay curves in the same sub-category energy decay model, then +.>、/>、/>All refer to attenuation coefficients of the subcategory under different energies, and the subcategory can be used as a subcategory to be verified of an object to be tested.

It should be noted that, the objects to be measured in S501 to S503 refer to single objects, that is, objects composed of one element, and for complex objects to be measured composed of multiple single objects, the method may be used to perform fitting matching on each single object in the complex objects to be measured, so as to obtain sub-categories to be verified of each single object in the complex objects to be measured.

For example, the to-be-tested object with the two kinds of simple substance materials overlapped is taken as an example, for the simple substance object 1 and the simple substance object 2 in the to-be-tested object, an actual energy attenuation curve of the simple substance object 1 under different energies and an actual energy attenuation curve of the simple substance object 2 under different energies can be obtained respectively, and for the simple substance object 1 and the simple substance object 2, the to-be-tested subcategory of the simple substance object 1 and the to-be-tested subcategory of the simple substance object 2 in the to-be-tested object can be determined by using the methods in the steps of S501 to S503.

Fig. 12 is a schematic flow chart of determining a target sub-category according to an embodiment of the present application, as shown in fig. 12, the determining the target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data in S105 may include:

s601, determining the density to be verified and the volume to be verified according to the subcategories to be verified.

Optionally, the density corresponding to the simple substance object can be known according to the sub-category to be verified, the volume to be verified is the product of the thicknesses obtained by scanning at different angles, and the density to be verified can be usedTo express, to be testedThe syndrome volume may be。

S602, determining the weight to be verified of the object to be tested according to the density to be verified and the volume to be verified.

In particular, the weight to be verified of the object to be tested may be the product of the density to be verified and the volume to be verified,。

s603, determining the target subcategory of the object to be tested according to the weight to be verified and the actual weight.

The above-mentioned S601 to S603 are weights to be verified for the elemental objects, and for a complex object to be tested composed of a plurality of elemental objects, for example, the complex object is composed of elemental object 1, elemental object 2, and elemental object 3, where the weights to be verified for elemental object 1 are The weight to be verified of the elemental object 2 is +.>The weight to be verified of the simple substance 3 isThe weight to be verified of the complex object to be tested is +.>。

Optionally, determining the target subcategory of the object to be tested according to the weight to be verified and the actual weight in S603 may include:

and if the difference value between the weight to be verified and the actual weight is smaller than the preset threshold value, taking the subcategory to be verified as the target subcategory.

The preset threshold may be set to 5%, or may be set to another value, which is not limited herein. When the difference between the weight to be verified and the actual weight is smaller than the preset threshold, it can be stated that the sub-category to be verified detected in the detection system is correct, and the sub-category to be verified can be used as the target sub-category, so that the post-processing system classifies the object to be detected according to the target sub-category.

If the difference value between the weight to be verified and the actual weight is greater than or equal to a preset threshold value, a re-detection instruction is sent to the visible light detection device and the X-ray detection device, new visible light detection data sent by the visible light detection device after re-detection and new multi-energy detection data sent by the X-ray detection device after re-detection are received, and the target subcategory of the object to be detected is re-determined according to the new visible light detection data and the new multi-energy detection data.

Fig. 13 is a schematic diagram of an apparatus for an object detection processing method according to an embodiment of the present application, where, as shown in fig. 13, the apparatus includes:

the receiving module 701 is configured to receive visible light detection data of an object to be detected sent by a visible light detection device, where the visible light detection data includes at least one of the following: shape information, actual weight, color;

a determining module 702, configured to determine a target class of the object to be detected according to the visible light detection data;

the receiving module 701 is configured to receive multi-energy detection data of the object to be detected sent by a multi-energy X-ray detection device, where the multi-energy detection data includes: the input energy, the output energy and the thickness of the object to be measured under various energies and various angles;

the obtaining module 703 is configured to obtain, according to the target category, a plurality of sub-category energy attenuation models corresponding to the target category, where each sub-category energy attenuation model is used to characterize a transformation condition of output energy of an object of each sub-category under the target category with thickness;

the determining module 702 is configured to determine a sub-category to be verified of the object to be tested according to the multi-energy detection data and the multiple sub-category energy attenuation models, and determine a target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data of the object to be tested.

Optionally, the determining module 702 is specifically configured to:

Fig. 14 is a block diagram of an electronic device 800 according to an embodiment of the present application. As shown in fig. 14, the electronic device may include: a processor 801, and a memory 802.

Optionally, a bus 803 may be further included, where the memory 802 is configured to store machine readable instructions executable by the processor 801, where the processor 801 communicates with the memory 802 via the bus 803 when the electronic device 800 is running, where the machine readable instructions are executed by the processor 801 to perform the method steps in the method embodiments described above.

The embodiment of the application also provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to execute the method steps in the embodiment of the object detection processing method.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described system and apparatus may refer to corresponding procedures in the method embodiments, and are not repeated in the present disclosure. In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, and the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, and for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, indirect coupling or communication connection of devices or modules, electrical, mechanical, or other form.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily appreciate variations or alternatives within the scope of the present application.

Claims

1. An object detection processing method, characterized in that the method comprises:

according to the target category, a plurality of sub-category energy attenuation models corresponding to the target category are obtained, and each sub-category energy attenuation model is used for representing the transformation condition of output energy of objects of each sub-category in the target category in the multi-energy detection data along with thickness;

determining a sub-category to be verified of the object to be tested according to the multi-energy detection data and the sub-category energy attenuation models, and determining a target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data of the object to be tested;

The determining the sub-category to be verified of the object to be tested according to the multi-energy detection data and the sub-category energy attenuation models comprises the following steps:

2. The object detection processing method according to claim 1, wherein the determining the target class of the object to be detected based on the visible light detection data includes:

3. The object detection processing method according to claim 1, wherein the constructing an energy attenuation curve of the object under test at each X-ray energy according to the input energy, the output energy, and the thickness of the object under test at a plurality of angles at each X-ray energy comprises:

and constructing an actual energy attenuation curve of the object to be measured under each X-ray energy according to the multi-angle output energy and thickness of the object to be measured under the current X-ray energy and the attenuation coefficient under the current X-ray energy.

4. The method according to claim 3, wherein determining the attenuation coefficient corresponding to each thickness of the object under test under the current X-ray energy according to the input energy, the output energy and the thickness of the object under test under the current X-ray energy at a plurality of angles comprises:

5. The method according to claim 1, wherein determining the sub-category to be verified of the object to be detected based on the actual energy attenuation curve of the object to be detected at each X-ray energy and the plurality of sub-category energy attenuation models comprises:

6. The method according to claim 1, wherein the determining the target sub-category of the object to be tested according to the sub-category to be verified includes:

7. The object detection processing method according to claim 6, wherein the determining the target sub-category of the object to be detected based on the weight to be verified and the actual weight includes:

8. An object detection processing apparatus, comprising:

the acquisition module is used for acquiring a plurality of sub-category energy attenuation models corresponding to the target category according to the target category, wherein each sub-category energy attenuation model is used for representing the conversion condition of the output energy of each sub-category object in the multi-energy detection data along with the thickness under the target category;

the determining module is used for determining a sub-category to be verified of the object to be tested according to the multi-energy detection data and the multiple sub-category energy attenuation models, and determining a target sub-category of the object to be tested according to the sub-category to be verified and the visible light detection data of the object to be tested;

The determining module is specifically configured to:

9. An object detection processing system, comprising: a visible light detection device, a multi-energy X-ray detection device, a classification device and a cloud server;

the cloud server is in communication connection with the visible light detection device, the multi-energy X-ray detection device and the classification device, and detects an object to be detected based on the method of any one of claims 1-7.

10. An electronic device comprising a memory and a processor, the memory storing a computer program executable by the processor, the processor implementing the steps of the object detection processing method of any of the preceding claims 1-7 when the computer program is executed.

11. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the object detection processing method according to any one of claims 1 to 7.