CN109934870B - Target detection method, device, equipment, computer equipment and storage medium - Google Patents
Target detection method, device, equipment, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109934870B CN109934870B CN201910091055.0A CN201910091055A CN109934870B CN 109934870 B CN109934870 B CN 109934870B CN 201910091055 A CN201910091055 A CN 201910091055A CN 109934870 B CN109934870 B CN 109934870B
- Authority
- CN
- China
- Prior art keywords
- target
- image
- target image
- area
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 103
- 238000000034 method Methods 0.000 claims abstract description 53
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 51
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims description 33
- 230000000877 morphologic effect Effects 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000004931 aggregating effect Effects 0.000 claims description 5
- 238000007621 cluster analysis Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The application relates to a target detection method, a target detection device, a target detection equipment, a computer equipment and a storage medium. The method comprises the following steps: the method comprises the steps of obtaining a target image, carrying out first target detection on the target image by adopting a maximum entropy and area combination algorithm, and carrying out second target detection switching processing on the target image by adopting a structural element top cap algorithm to obtain a first target or a second target in the target image. The method can improve the detection accuracy and reduce the false alarm rate.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a target detection method, apparatus, device, computer device, and storage medium.
Background
With the development of image processing techniques, algorithms have emerged to detect objects of different sizes in images. The method includes a method using a multi-scale space theory, a method related to machine learning, and the like.
The principle of the method of the multi-scale space theory is complex, the adaptability to the scene in the actual scene is not high, and the instantaneity is not easy to achieve. For machine learning-related based methods, this class of methods requires a large amount of training data to train the classifier, but what the target detection study lacks is the data.
However, the current method has the problems of high false alarm rate and the like.
Disclosure of Invention
In view of the above, it is necessary to provide an object detection method, apparatus, device, computer device and storage medium for solving the above technical problems.
A method of target detection, the method comprising:
an image of the object is acquired,
and carrying out first target detection on the target image by adopting a maximum entropy and area combination algorithm and carrying out second target detection switching processing on the target image by adopting a structural element top hat algorithm so as to obtain a first target or a second target in the target image.
In one embodiment, the performing, by using a maximum entropy and area combination algorithm, a first target detection on the target image and performing, by using a structural element top hat algorithm, a second target detection switching process on the target image to obtain a first target or a second target in the target image includes:
and if the first target is not detected in at least one continuous frame, performing second target detection on the target image by adopting a structural element top hat algorithm.
In one embodiment, the second target detection of the target image by using the structural element top hat algorithm if the first target is not detected in at least one consecutive frame comprises:
and if the second target is not detected in at least one continuous frame, returning to the step of detecting the first target in the target image.
In one embodiment, the performing the first target detection on the target image by using the maximum entropy and area combination algorithm includes:
acquiring a threshold value for segmenting the target image, and segmenting the target image according to the threshold value to obtain a first image;
performing connected domain analysis on the first image to obtain a second image;
and carrying out contrast analysis on the second image to obtain the first target.
In one embodiment, the obtaining a threshold for segmenting the target image, and segmenting the target image according to the threshold to obtain a first image includes:
acquiring the gray scale of the target image, and respectively calculating the gray entropy of the target image corresponding to the gray scale and the area of the target image corresponding to the gray scale according to the gray scale, wherein the area of the target image comprises a target area and a background area;
acquiring a region area difference of the target image according to the area of the target region and the area of a background region;
and if the product value of the gray level entropy of the target image and the area difference of the target image is maximum, acquiring a threshold value corresponding to the gray level when the product value is maximum.
In one embodiment, the performing connected component analysis on the first image to obtain a second image includes:
and aggregating the target areas in the first image by adopting cluster analysis to obtain the selectable target area.
In one embodiment, the performing contrast analysis on the second image to obtain the first target includes:
and removing non-target areas in the selectable target area by adopting contrast analysis to obtain the first target.
In one embodiment, the performing, by using the structural element top hat algorithm, the second target detection on the target image includes:
constructing a morphological structure element of the target image, and performing morphological top hat operation on the target image according to the morphological structure element to obtain a third image;
subtracting the third image from the target image to obtain a fourth image;
acquiring a second threshold value, and segmenting the fourth image to obtain a fifth image;
performing connected domain analysis on the fifth image to obtain a sixth image;
and carrying out contrast analysis on the sixth image to obtain the second target.
In one embodiment, the method further comprises:
and acquiring an initial image, and preprocessing the initial image to obtain the target image.
In one embodiment, the acquiring an initial image and preprocessing the initial image to obtain a target image includes:
and sequentially carrying out Gaussian filtering smoothing and maximum value filtering denoising treatment on the initial image to obtain the target image.
An object detection apparatus, the apparatus comprising:
and the target acquisition module is used for acquiring a target image, performing first target detection on the target image by adopting a maximum entropy and area combination algorithm and performing second target detection switching processing on the target image by adopting a structural element top hat algorithm so as to obtain a first target or a second target in the target image.
An object detection apparatus, said apparatus comprising an object detection device.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method as claimed in any one of the above when the computer program is executed.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of the preceding claims.
According to the target detection method, the device, the equipment, the computer equipment and the storage medium, the first target detection is carried out on the target image by acquiring the target image and adopting a maximum entropy and area combination algorithm, and the second target detection switching processing is carried out on the target image by adopting a structural element top cap algorithm, so that the first target or the second target in the target image is obtained. By the method, the detection accuracy can be improved, and the false alarm rate can be reduced.
Drawings
FIG. 1 is a diagram of an exemplary implementation of a target detection method;
FIG. 2 is a schematic flow chart diagram illustrating a method for target detection in one embodiment;
FIG. 3 is a flowchart illustrating step S1 according to an embodiment;
FIG. 4 is a flowchart illustrating step S12 according to another embodiment;
FIG. 5 is a flowchart illustrating step S121 in another embodiment;
FIG. 6 is a flowchart illustrating step S11 according to another embodiment;
FIG. 7 is a schematic flow chart of a target detection method in another embodiment;
FIG. 8(a) is a detected image of a large object of a scene;
FIG. 8(b) is a detected image of a small object of a scene;
FIG. 8(c) is a detected image of two large objects in a scene;
FIG. 8(d) is a detected image of two small objects in a scene;
fig. 9(a) is a detection image of a large target in a medium complex sky background;
FIG. 9(b) is a detected image of a small target in a medium complex sky background;
fig. 10(a) is a detection image of a large target in a background of a complex sky;
fig. 10(b) is a detection image of a small target in a background of a complex sky;
FIG. 11 is a block diagram of an object detection device in one embodiment;
FIG. 12 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The target detection method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The terminal 102 acquires a target image, transmits the target image to the server 104, and the server 104 processes the target image as follows: and carrying out first target detection on the target image by adopting a maximum entropy and area combination algorithm and carrying out second target detection switching processing on the target image by adopting a structural element top hat algorithm so as to obtain a first target or a second target in the target image. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, an object detection method is provided, which is described by taking the application of the method to the server in fig. 1 as an example, and includes the following steps:
a method of target detection, the method comprising:
step S1: an image of the object is acquired,
and carrying out first target detection on the target image by adopting a maximum entropy and area combination algorithm and carrying out second target detection switching processing on the target image by adopting a structural element top hat algorithm so as to obtain a first target or a second target in the target image.
Specifically, the target image refers to an image having a target and a complex background. The complex background at least comprises a sky background or a sea background and the like. Targets refer to infrared targets, including first targets and second targets, where the first targets have different sizes than the second targets, e.g., 9 x 9 objects for the first targets and 19 x 19 objects for the second targets.
Performing first target detection on the target image by adopting a maximum entropy and area combination algorithm and performing second target detection switching processing on the target image by adopting a structural element top hat algorithm, namely if the first target is not detected when the first target detection is performed on the target image by adopting the maximum entropy and area combination algorithm, performing second target detection on the target image by adopting the structural element top hat algorithm until the first target or the second target is detected; or, if the second target is not detected when the structural element top hat algorithm is adopted to detect the second target of the target image, the first target detection is carried out on the target image by adopting the maximum entropy and area combination algorithm until the first target or the second target is detected. Wherein the above execution order may be any.
According to the method and the device, the strategy of switching the first target detection algorithm and the second target detection algorithm is used, the problem of target detection of infrared targets with different sizes under various high-altitude backgrounds is solved, the detection rate of the infrared targets is improved, and the adaptability of the whole product is enhanced. By adopting a maximum entropy and area combination algorithm, the method has good adaptability to the target scene of a complex cloud layer, can effectively inhibit the interference of backgrounds such as the cloud layer and the like, improves the detection rate of target detection, and well reduces the false scene rate through subsequent connected domain analysis and contrast analysis; by adopting the structural element top hat algorithm, targets in simple and complex sky scenes can be effectively detected, a large background can be effectively suppressed, and the false scene rate is well reduced by a connected domain analysis and contrast analysis method.
According to the target detection method, a target image is obtained, first target detection is carried out on the target image by adopting a maximum entropy and area combination algorithm, and second target detection switching processing is carried out on the target image by adopting a structural element top cap algorithm, so that a first target or a second target in the target image is obtained. By the method, the detection accuracy can be improved, and the false alarm rate can be reduced.
In one embodiment, with reference to fig. 3, the step S1 includes:
step S11: and if the first target is not detected in at least one continuous frame, performing second target detection on the target image by adopting a structural element top hat algorithm.
Specifically, when the first target detection is performed on the acquired target image, the value range of the frame number is 3 to 5.
In one embodiment, the step S11 is followed by:
step S12: and if the second target is not detected in at least one continuous frame, returning to the step of detecting the first target in the target image.
Specifically, at least one frame refers to n frames, n is a positive integer, and n > 1. When the second target detection is performed on the acquired target image, the value range of the frame number can also be 3-5, which is the same as the value of the frame number set when the first target is detected.
In one embodiment, with reference to fig. 4, the step S12 includes:
step S121: and acquiring a threshold value for segmenting the target image, and segmenting the target image according to the threshold value to obtain a first image.
Specifically, the first image refers to an image obtained by segmenting the target image by a threshold value.
Step S122: and analyzing the connected domain of the first image to obtain a second image.
Specifically, connected component analysis refers to the formation of different regions according to the division of the binary image into regions. The dots which are connected with each other in the binary image form a region, the dots which are not connected with each other form different regions, and a set of all the dots which are connected with each other is called a connected region.
Connected component analysis refers to aggregating discrete points in the image into image blocks. The second image is an image obtained by analyzing the connected domain of the first image.
Step S123: and carrying out contrast analysis on the second image to obtain the first target.
Specifically, the purpose of the contrast analysis is to remove the false alarm in the second image, preserving the target in the image.
In one embodiment, with reference to fig. 5, the step S121 includes:
step S1211: acquiring the gray scale of the target image, and respectively calculating the gray entropy of the target image corresponding to the gray scale and the area of the target image corresponding to the gray scale according to the gray scale, wherein the area of the target image comprises a target area and a background area;
step S1212: acquiring a region area difference of the target image according to the area of the target region and the area of a background region;
step S1213: and if the product value of the gray level entropy of the target image and the area difference of the target image is maximum, acquiring a threshold value corresponding to the gray level when the product value is maximum.
Let L be the total number of gray levels (gray levels) of an image, g (k) be the number of pixels with k gray levels in the image, k being 0,1, …L-1, dividing the image into target classes (target area) CoAnd a background class (background region) CbFor convenience of discussion, if the low gray scale region is set as the target class and the high gray scale region is set as the background class, then:
the image gray entropy is:
wherein Ho(t) is a target class gray level entropy; hbAnd (t) is background class gray entropy. The image gray entropy represents the macroscopic statistical characteristics of image energy distribution and reflects the difference degree of pixel gray in the region. The larger the image gray entropy is, the smaller the pixel gray difference in the class is, and when the gray entropy reaches the maximum, the gray of the target class and the background class tends to be uniform.
In addition, considering that the infrared target area is usually much smaller than the background area, the failure of target segmentation is easily caused by directly adopting threshold segmentation, and a threshold selection formula is constructed by utilizing the characteristic that the area difference between the target area and the background area is very large.
Setting the area of the target region as So(t) the area of the background region is Sb(t), then:
the final threshold selection function is:
η(t)＝H(t)·[So(t)-Sb(t)]2；
t when η (t) takes the maximum value is the optimal threshold:
in one embodiment, the step S122 includes:
step S1221: and aggregating the target areas in the first image by adopting cluster analysis to obtain the selectable target area.
Specifically, the selectable target area includes a target area and a suspected target area, and the suspected target area is excluded through post-processing to obtain the target area so as to identify the target.
In one embodiment, the step S123 includes:
step S1231: and removing non-target areas in the selectable target area by adopting contrast analysis to obtain the first target.
Specifically, through contrast analysis, a virtual scene in a suspected target area is excluded, and target display is reserved.
The contrast ratio is formulated as follows:
LCM＝Tmax-Bmean
wherein T ismaxIs the maximum value of the target area, BmeanIs the mean of the background area. And after the contrast is calculated, taking the standard deviation of the mean value plus a certain weight as a threshold value for all the contrasts to segment and display the target area.
In one embodiment, with reference to fig. 6, the step S11 includes:
step S111: and constructing a morphological structure element of the target image, and performing morphological top hat operation on the target image according to the morphological structure element to obtain a third image.
Specifically, the third image refers to an image obtained after the morphological top hat operation is performed on the object. And constructing the structural elements according to the smaller target of the first target and the second target, wherein the second target is the smaller target and has the size of 9 x 9, constructing the special structural elements with the middle value of 9 x 9 of 0 and the rest value of 19 x 19 of 1, and performing morphological top hat operation to perform morphological expansion and corrosion on the image.
Step S112: and subtracting the third image from the target image to obtain a fourth image.
Specifically, the third image is subtracted from the target image to obtain a fourth image, wherein the fourth image includes the target and the interference information.
Step S113: and acquiring a second threshold value, and segmenting the fourth image to obtain a fifth image.
Specifically, the second threshold refers to a threshold required for segmenting the fourth image, and the threshold is set as a standard deviation of an image mean plus a certain weight. The fifth image is an image obtained by dividing the fourth image according to a second threshold.
In order to adapt to different complex scene changes, the threshold value is adaptively selected according to the statistical information of the contrast value. The threshold is calculated as follows:
wherein,
the mean and standard deviation of the image are respectively, k is a constant and ranges from 1 to 2, and the method is set to be 1.5 according to the actual scene.
Step S114: and analyzing the connected domain of the fifth image to obtain a sixth image.
Specifically, when the connected domain analysis is performed on the fifth image, a clustering method is mainly adopted to aggregate points in the fifth image into image blocks so as to obtain a suspected target area. And the sixth image is obtained after connected domain analysis is carried out on the fifth image, wherein the sixth image comprises a suspected target area.
Step S115: and carrying out contrast analysis on the sixth image to obtain the second target.
Specifically, the method comprises the following steps: and through contrast analysis, false alarms in the suspected target area are eliminated, and target display is reserved.
The contrast ratio is formulated as follows:
wherein T ismaxIs the maximum value of the target area, BmaxIs the maximum value of the background region, BminIs the minimum of the background area. And after the contrast is calculated, taking the standard deviation of the mean value plus a certain weight as a threshold value for all the contrasts to segment and display the target area.
In one embodiment, with reference to fig. 7, the method further includes:
step S2: and acquiring an initial image, and preprocessing the initial image to obtain the target image.
Specifically, image preprocessing is to eliminate irrelevant information from the original image, recover useful real information, enhance the detectability of relevant information and simplify the data to the maximum extent, thereby improving the reliability of feature extraction, image segmentation, matching and recognition.
In one embodiment, the step S2 includes:
step S21: and sequentially carrying out Gaussian filtering smoothing and maximum value filtering denoising treatment on the initial image to obtain the target image.
Specifically, gaussian filtering is performed through the 3 × 3 template to remove gaussian noise, and then maximum filtering is performed through the 3 × 3 template to facilitate subsequent contrast calculation.
In one embodiment, the following is the target detection results using infrared data of different complexity sky backgrounds in actual testing.
FIG. 8 is a diagram of two relatively simple infrared target detections in a background scene of the sky, where FIG. 8(a) is a detected image of a large target in the scene and FIG. 8(b) is a detected image of a small target in the scene; fig. 8(c) is a detection image of a large object in the scene two, and fig. 8(d) is a detection image of a small object in the scene two.
Through the image, the infrared target in the image can be well detected by the algorithm.
Fig. 9 shows the detection results of infrared targets with different sizes in the background of medium and complex sky, and it can be seen that the algorithm of the present application can well detect the infrared targets in the image.
Fig. 9(a) is a detection image of a large target in a medium-complex sky background, and fig. 9(b) is a detection image of a small target in a medium-complex sky background.
Fig. 10 shows the detection results of infrared targets with different sizes in a complex sky background, and it can be seen that the algorithm of the present application can well detect infrared targets in an image.
Fig. 10(a) is a detection image of a large target in a background of a complex sky, and fig. 10(b) is a detection image of a small target in a background of a complex sky.
The observation of the results shows that the method provided by the application can detect the infrared targets with different scales under the sky backgrounds with different complexities, and can effectively detect the targets and reduce the false scene rate. In an actual product, the FPGA and DSP framework is used, the switching between the surface target and the small target is used, the real-time requirement can be effectively met under the condition of not influencing the precision, and the effectiveness and the robustness of the method are explained.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 11, there is provided an object detection apparatus including: a target acquisition module, wherein:
the target obtaining module 10 is configured to obtain a target image, perform first target detection on the target image by using a maximum entropy and area combination algorithm, and perform second target detection switching processing on the target image by using a structural element top hat algorithm, so as to obtain a first target or a second target in the target image.
In one embodiment, the object acquisition module 10 includes:
and the second target detection module 101 is configured to perform second target detection on the target image by using a structural element top hat algorithm if the first target is not detected in at least one continuous frame.
In one embodiment, the second object detection module 101 comprises, after:
the first target detection module 102 is configured to, if the second target is not detected in at least one consecutive frame, return to the step of performing first target detection on the target image.
In one embodiment, the first object detection module 102 includes:
a first image obtaining module 1021, configured to obtain a threshold for segmenting the target image, and segment the target image according to the threshold to obtain a first image;
a second image obtaining module 1022, configured to perform connected domain analysis on the first image to obtain a second image;
the first target obtaining module 1023 is configured to perform contrast analysis on the second image to obtain the first target.
In one embodiment, the first image acquisition module 1021 includes:
the calculation module 1021 a: acquiring the gray scale of the target image, and respectively calculating the gray entropy of the target image corresponding to the gray scale and the area of the target image corresponding to the gray scale according to the gray scale, wherein the area of the target image comprises a target area and a background area;
the area difference obtaining module 1021 b: acquiring a region area difference of the target image according to the area of the target region and the area of a background region;
threshold acquisition module 1021 c: and if the product value of the gray level entropy of the target image and the area difference of the target image is maximum, acquiring a threshold value corresponding to the gray level when the product value is maximum.
In one embodiment, the second image acquisition module 1022 includes:
the optional target area acquisition module 1022 a: and aggregating the target areas in the first image by adopting cluster analysis to obtain the selectable target area.
In one embodiment, the first object acquisition module 1023 comprises:
non-target region culling module 1023 a: and removing non-target areas in the selectable target area by adopting contrast analysis to obtain the first target.
In one embodiment, the second object detection module 101 comprises:
a third image obtaining module 1011, configured to construct a morphological structural element of the target image, and perform a morphological top hat operation on the target image according to the morphological structural element to obtain a third image;
a fourth image obtaining module 1012, configured to subtract the third image from the target image to obtain a fourth image;
a fifth image obtaining module 1013, configured to obtain a second threshold, and segment the fourth image to obtain a fifth image;
a sixth image obtaining module 1014, configured to perform connected domain analysis on the fifth image to obtain a sixth image;
sixth graphics processing module 1014 a: and carrying out contrast analysis on the sixth image to obtain the second target.
In one embodiment, the method further comprises:
the preprocessing module 20 is configured to acquire an initial image, and preprocess the initial image to obtain the target image.
In one embodiment, the preprocessing module 20 includes:
and the target image acquisition module 21 is configured to perform gaussian filtering smoothing and maximum filtering denoising processing on the initial image in sequence to obtain the target image.
In one embodiment, there is also provided an object detection apparatus comprising an object detection device.
For a specific definition of the target detection device, reference may be made to the above definition of a target detection method, which is not described herein again. The modules in the target detection device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 12. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing object detection data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of object detection.
Those skilled in the art will appreciate that the architecture shown in fig. 12 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, a computer device is provided, comprising a memory in which a computer program is stored and a processor which, when executing the computer program, carries out the steps as described in the above method.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method as described above.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (13)
1. A method of object detection, the method comprising:
acquiring a target image, and performing first target detection on the target image by adopting a maximum entropy and area combination algorithm and performing second target detection switching processing on the target image by adopting a structural element top cap algorithm to obtain a first target or a second target in the target image;
wherein, the determination mode of the morphological structural element in the structural element top cap algorithm comprises the following steps: constructing the morphological structural element according to the smaller of the first and second objects;
the performing, by using a maximum entropy and area combination algorithm, first target detection on the target image and performing, by using a structural element top hat algorithm, second target detection switching processing on the target image to obtain a first target or a second target in the target image includes:
and if the first target is not detected in at least one continuous frame, performing second target detection on the target image by adopting a structural element top hat algorithm.
2. The method of claim 1, wherein after performing a second object detection on the object image by using a structural element top hat algorithm if the first object is not detected in at least one consecutive frame, the method comprises:
and if the second target is not detected in at least one continuous frame, returning to the step of detecting the first target in the target image.
3. The method of claim 1, wherein the first target detection of the target image using a maximum entropy and area combination algorithm comprises:
acquiring a threshold value for segmenting the target image, and segmenting the target image according to the threshold value to obtain a first image;
performing connected domain analysis on the first image to obtain a second image;
and carrying out contrast analysis on the second image to obtain the first target.
4. The method of claim 3, wherein obtaining a threshold for segmenting the target image, and segmenting the target image according to the threshold to obtain the first image comprises:
acquiring the gray scale of the target image, and respectively calculating the gray entropy of the target image corresponding to the gray scale and the area of the target image corresponding to the gray scale according to the gray scale, wherein the area of the target image comprises a target area and a background area;
acquiring a region area difference of the target image according to the area of the target region and the area of a background region;
and if the product value of the gray level entropy of the target image and the area difference of the target image is maximum, acquiring a threshold value corresponding to the gray level when the product value is maximum.
5. The method of claim 4, wherein performing connected component analysis on the first image to obtain a second image comprises:
and aggregating the target areas in the first image by adopting cluster analysis to obtain the selectable target area.
6. The method of claim 5, wherein performing a contrast analysis on the second image to obtain the first target comprises:
and removing non-target areas in the selectable target area by adopting contrast analysis to obtain the first target.
7. The method of claim 2, wherein the second target detection of the target image using the structural element top hat algorithm comprises:
performing morphological top hat operation on the target image according to the morphological structural element to obtain a third image;
subtracting the third image from the target image to obtain a fourth image;
acquiring a second threshold value, and segmenting the fourth image to obtain a fifth image;
performing connected domain analysis on the fifth image to obtain a sixth image;
and carrying out contrast analysis on the sixth image to obtain the second target.
8. The method of claim 1, further comprising:
and acquiring an initial image, and preprocessing the initial image to obtain the target image.
9. The method of claim 8, wherein the obtaining an initial image and pre-processing the initial image to obtain a target image comprises:
and sequentially carrying out Gaussian filtering smoothing and maximum value filtering denoising treatment on the initial image to obtain the target image.
10. An object detection apparatus, characterized in that the apparatus comprises:
the target acquisition module is used for acquiring a target image, performing first target detection on the target image by adopting a maximum entropy and area combination algorithm and performing second target detection switching processing on the target image by adopting a structural element top hat algorithm to obtain a first target or a second target in the target image;
wherein, the determination mode of the morphological structural element in the structural element top cap algorithm comprises the following steps: constructing the morphological structural element according to the smaller of the first and second objects;
the target acquisition module includes: and the second target detection module is used for performing second target detection on the target image by adopting a structural element top hat algorithm if the first target is not detected in at least one continuous frame.
11. An object detection device, characterized in that it comprises the apparatus of claim 10.
12. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 9 when executing the computer program.
13. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910091055.0A CN109934870B (en) | 2019-01-30 | 2019-01-30 | Target detection method, device, equipment, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910091055.0A CN109934870B (en) | 2019-01-30 | 2019-01-30 | Target detection method, device, equipment, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109934870A CN109934870A (en) | 2019-06-25 |
| CN109934870B true CN109934870B (en) | 2021-11-30 |
Family
ID=66985319
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910091055.0A Active CN109934870B (en) | 2019-01-30 | 2019-01-30 | Target detection method, device, equipment, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109934870B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418200B (en) * | 2021-01-25 | 2021-04-02 | 成都点泽智能科技有限公司 | Object detection method and device based on thermal imaging and server |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330857A (en) * | 2017-06-21 | 2017-11-07 | 中国科学院长春光学精密机械与物理研究所 | Dark dim point-target detection method in a kind of infrared image |
| EP3353706A1 (en) * | 2015-09-15 | 2018-08-01 | SZ DJI Technology Co., Ltd. | System and method for supporting smooth target following |
| CN110223344A (en) * | 2019-06-03 | 2019-09-10 | 哈尔滨工程大学 | A kind of infrared small target detection method based on morphology and vision noticing mechanism |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016015251A1 (en) * | 2014-07-30 | 2016-02-04 | SZ DJI Technology Co., Ltd. | Systems and methods for target tracking |
| CN104268844B (en) * | 2014-10-17 | 2017-01-25 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
| CN106934819A (en) * | 2017-03-10 | 2017-07-07 | 重庆邮电大学 | A kind of method of moving object segmentation precision in raising image |
| CN107609498B (en) * | 2017-09-01 | 2020-08-18 | 湖州师范学院 | Data processing method of computer monitoring system |
| CN108445480B (en) * | 2018-02-02 | 2022-05-03 | 重庆邮电大学 | Mobile platform self-adaptive extended target tracking system and method based on laser radar |
-
2019
- 2019-01-30 CN CN201910091055.0A patent/CN109934870B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3353706A1 (en) * | 2015-09-15 | 2018-08-01 | SZ DJI Technology Co., Ltd. | System and method for supporting smooth target following |
| CN107330857A (en) * | 2017-06-21 | 2017-11-07 | 中国科学院长春光学精密机械与物理研究所 | Dark dim point-target detection method in a kind of infrared image |
| CN110223344A (en) * | 2019-06-03 | 2019-09-10 | 哈尔滨工程大学 | A kind of infrared small target detection method based on morphology and vision noticing mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109934870A (en) | 2019-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109271870B (en) | Pedestrian re-identification method, device, computer equipment and storage medium | |
| CN108765369B (en) | Method, apparatus, computer device and storage medium for detecting lung nodule | |
| CN111860670A (en) | Domain adaptive model training method, image detection method, device, equipment and medium | |
| CN108764325B (en) | Image recognition method and device, computer equipment and storage medium | |
| CN110059700B (en) | Image moire recognition method and device, computer equipment and storage medium | |
| CN113239874B (en) | Behavior gesture detection method, device, equipment and medium based on video image | |
| CN111368758B (en) | Face ambiguity detection method, face ambiguity detection device, computer equipment and storage medium | |
| CN111862044B (en) | Ultrasonic image processing method, ultrasonic image processing device, computer equipment and storage medium | |
| CN109903272B (en) | Target detection method, device, equipment, computer equipment and storage medium | |
| CN110569721A (en) | Recognition model training method, image recognition method, device, equipment and medium | |
| CN110532851B (en) | Finger vein identification method and device, computer equipment and storage medium | |
| CN112560796B (en) | Human body posture real-time detection method and device, computer equipment and storage medium | |
| CN112183252B (en) | Video motion recognition method, device, computer equipment and storage medium | |
| CN109035287B (en) | Foreground image extraction method and device and moving vehicle identification method and device | |
| CN114155365B (en) | Model training method, image processing method and related device | |
| CN112464829B (en) | Pupil positioning method, pupil positioning equipment, storage medium and sight tracking system | |
| WO2022121130A1 (en) | Power target detection method and apparatus, computer device, and storage medium | |
| CN112232140A (en) | Crowd counting method and device, electronic equipment and computer storage medium | |
| CN112348116A (en) | Target detection method and device using spatial context and computer equipment | |
| CN111445487A (en) | Image segmentation method and device, computer equipment and storage medium | |
| CN109887001A (en) | Method for tracking target, device, computer equipment and storage medium | |
| CN113537020A (en) | Complex SAR image target identification method based on improved neural network | |
| CN109934870B (en) | Target detection method, device, equipment, computer equipment and storage medium | |
| WO2020062546A1 (en) | Target tracking processing method and electronic device | |
| CN114092515A (en) | Target tracking detection method, device, equipment and medium for obstacle blocking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |