WO2015040626A1

WO2015040626A1 - Quantitative analysis of milk components

Info

Publication number: WO2015040626A1
Application number: PCT/IN2013/000686
Authority: WO
Inventors: Fredrick. H. LONG; Ajit PATEL
Original assignee: Everest Instruments Pvt. Ltd.
Priority date: 2013-09-20
Filing date: 2013-11-07
Publication date: 2015-03-26
Also published as: IN2013MU03035A

Abstract

A milk analyzer for conducting quantitative analysis of milk components of test milk sample, for which the quantities of component parts of the test milk sample are unknown. The milk analyzer (100) includes a fiber-optic probe (102) implemented in a diffuse reflectance mode and a near-infrared spectrometer (108). The fiber-optic probe (102) transmits a light in a preselected wavelength range between 600 nanometers (nm) and 1100 into the test milk sample and receives a reflected light from the test milk sample. The received reflected light is provided to the NIR spectrometer (108) to perform spectral analysis of the reflected light for determining reflectance spectra of the test milk sample. The milk analyzer (100) further includes a controller (110) for conducting quantitative analysis of the test milk samples based on the determined reflectance spectra and a predetermined single reference model to determine quantities of milk components of the test milk sample.

Description

TECHNICAL FIELD

[0001] The present subject matter relates, in general, to analysis of milk, and particularly, but not exclusively to, quantitative analysis of milk components.

BACKGROUND

[0002] Raw milk collected from dairy farms is generally analyzed to quantify the component parts in the raw milk and to assess the quality of the raw milk. The quantitative analysis of raw milk includes measurement of quantities of milk components such as fat, solid non fat (SNF), proteins, minerals, and water, and other parameters such as acidity and bacteriological quality. The quantitative analysis of the milk helps in quality control and management of dairy farms. Milk is a strong light-scattering medium due to the presence of colloidal fat (1-10 μηι) and protein (80-200 nm) particles. Therefore, present day techniques of quantitative analysis of milk components usually involve spectroscopic analysis of milk in the near infrared (NIR) or mid infrared (MIR) wavelength ranges.

BRIEF DESCRIPTION OF DRAWINGS

[0003] The detailed description is described with reference to the accompanying figures.

In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to reference like features and components.

[0004] Fig. la illustrates a milk analyzer for conducting quantitative analysis of milk components, in accordance with an embodiment of the present subject matter.

[0005] Fig. lb and lc illustrate internal view of the milk analyzer showing various hardware components in the milk analyzer, in accordance with the embodiment Fig la.

[0006] Fig. 2a, 2b, and 2c illustrates experimental results of quantitative analysis of milk components, in accordance with an embodiment of the present subject matter.

[0007] Fig. 3 illustrates a method for conducting quantitative analysis of milk components, in accordance with an embodiment of the present subject matter. DETAILED DESCRIPTION

[0008] A milk analyzer for conducting quantitative analysis of milk components is described herein. Milk is an important source of nutrition especially for children and infants. Milk products also form a significant part of the diet of adults around the world. It is estimated that at present more than 6 billion consumers use milk and milk products. As such, milk is considered as an important factor in improving nutrition and reducing malnutrition. With advancement in technology, improvements have been seen in cattle breeding and dairy technology for obtaining larger quantities of milk. After the milk is collected from the cattle in a dairy farm, the milk is typically analyzed for quantifying milk components and for assessing quality of the milk. The quantitative analysis of milk includes measuring quantities of various milk components such as fat, solid non fat (SNF), proteins, minerals, water, and other parameters such as acidity and bacteriological activity. An assessment of the quantities of these milk components helps in herd management and dairy trade.

[0009] Further, different types of milks which are sold in the market are also processed and separated on the basis of the quantities of these milk components, especially fat and SNF. For instance, whole milk, which is considered best only for infants, contains at least 3.25% fat and 8.25% SNF by weight. The fat and SNF in the milk, among other milk components, are important indicators of the quality of milk and as such determine the price of the milk. The quantities of the fat and SNF are also used in analyzing and modifying diet of the cattle so that milk of standard quality can be obtained. For example, changes in the fat content may indicate an imbalance in the forage-to-concentrate ratio in the feed.

[0010] Traditional methods available for analyzing milk include chemical and infrared analysis, which are difficult to perform at a dairy farm. Some other techniques include spectroscopic analysis of the milk in the visible wavelength range. For example, Hunter Color Method is used for quantifying amount of fat present in a milk sample. However, these techniques use expensive instruments which cannot be readily used in the dairy farms. As a result, such instruments are confined to centralized laboratories, to which a farmer would periodically send milk samples for testing, say once a month.

[0011] Modern techniques offer on-line milk analysis using spectroscopic analysis at near infrared (NIR) wavelength range due to better accuracy of quantification obtained from such techniques. One such technique involves spectroscopic analysis of milk based on absorbance of light in the NIR range between 800 nm and 2500 nm. However, measuring absorbance at wavelengths above 1,100 nm requires specialized detectors, which are expensive, thereby increasing the overall cost of using the technique.

[0012] Further, the spectroscopic measurements are typically obtained using a commercial NIR spectroscopic analyzer, such as FOSS Rapid Content Analyzer™ (RCA) 6500. The measurements are then transferred to a computer, wherein the measurements are used for determining fat content based on an analysis of the measured spectra. However, the cost of such a commercial NIR spectroscopic milk analyzer is very high, and therefore not affordable by typical dairy farms. Moreover, the space constraints in the dairy farms make the installation of such analytical instruments difficult. Additionally, the present techniques rely on analytical models which are specific to one kind of milk, such as milk sourced from a particular breed of cattle, thereby necessitating use of different models for analyzing different kinds of milk.

[0013] According to an embodiment of the present subject matter, a milk analyzer for conducting quantitative analysis of milk components is described herein. The milk analyzer includes a light source for emitting light in a wavelength range between 600 nanometers (nm) and 1100 nm. The milk analyzer also includes a fiber-optic probe implemented in a diffuse reflectance mode for transmitting light emitted from the light source to interior of test milk sample, for which spectra and quantities of milk components are unknown. The light is transmitted by the fiber-optic probe in such a way that the light is incident on the interior of the test milk sample. The fiber-optic probe, being implemented in the diffused reflectance mode, receives a reflected light from the interior of the test milk sample. The milk analyzer further includes a near-infrared (NIR) spectrometer for receiving the reflected light through the fiberoptic probe and conducting spectroscopic analysis of the reflected light to determine a reflectance spectra. In one implementation, parameters of the NIR spectrometer, including data acquisition interval and data acquisition time, are set based on hardware components used in the NIR spectrometer. The values of the parameters are set such that the sampling and analysis of the received light can be performed with much faster rate over long intervals and without significant loss of information. Further, the parameters can be set at different values based on selection of affordable hardware components of the NIR spectrometer such as detectors. It will be understood that the parameters of the NIR spectrometer can be varied depending on, for example, a detector and grating combination used in the NIR spectrometer. [0014] The milk analyzer further includes a controller for conducting quantitative analysis of milk components based on the spectroscopic analysis of the reflected light and a display unit for displaying results of the quantitative analysis. In one implementation, the display unit can be an interactive display unit, such as a touch-screen display and can include user- interface for enabling a user to use the milk analyzer. For example, the user-interface can include various menu bars indicative of various functions performed by the milk analyzer such as performing self-calibration and initializing quantitative analysis. Further, in one implementation, the parameters of the NIR spectrometer are set through the user-interface in the display unit. Thus, the milk analyzer, according to the present subject matter, is a compact and portable analytical instrument having all hardware components, namely the NIR spectrometer, the light source, the fiber-optic probe, the controller, and the display unit, coupled together in one unit.

[0015] In an embodiment, the milk analyzer can be used for conducting quantitative analysis of milk components based on a single reference model. In an implementation, the analyzed milk components are fat and SNF. The single reference model is a regression model generated based on spectral analysis of a set of reference milk samples, for which spectra and quantities of milk components are known. The quantities of milk components of the set of reference milk samples can be measured using any known methods. The known spectra can be measured using same method used for performing spectral analysis of the set of reference milk samples. The set of reference milk samples can include cow milk, buffalo milk, mixed milk, and can also be milk from other animal sources, such as goat, sheep and camel. In one embodiment, the single reference model may be generated by an industrial user, which may include, but is not limited to, a research institute, analytical instruments manufacturing company, and the like. The generated single reference model can then be programmed in the milk analyzer and used for conducting on-line quantitative analysis of milk components of test milk samples for which spectra and quantities of milk components are unknown. In one implementation, the generated single reference model is saved in a memory of the milk analyzer.

[0016] The present subject matter thus, provides a milk analyzer for conducting quantitative analysis of milk components. The milk analyzer provides a single platform for obtaining and analyzing milk spectra for quantitative analysis by having all hardware components in one device, thereby making the milk analyzer a complete on-line, stand-alone analytical instrument for milk analysis. Thus, unlike commercially available NIR spectroscopic analyzers, external computer is not required to be coupled with the milk analyzer for transferring data and for conducting quantitative analysis. Therefore, the milk analyzer can obtain data and perform on-line quantitative analysis in a dairy process. Also, the display unit on the milk analyzer provides an interactive user-interface having various menu bars such that a user can easily use the milk analyzer, with minimal training, on the diary farm. Moreover, the fiber-optic probe is implemented in a diffuse reflectance mode, which makes the cleaning of the fiber-optic probe easier, thereby enabling analysis of a plurality of test milk samples using the same probe and reducing maintenance costs of the milk analyzer.

[0017] Further, the milk analyzer includes a low cost, lower resolution NIR spectrometer with much faster data acquisition rate, thus making the milk analyzer cheap and affordable, especially, for dairy farms and also providing faster analysis results. Furthermore, the milk analyzer does not require an external control system for performing the necessary analysis as the control system is built within the milk analyzer. Further, different types of test milk samples can be analyzed based on a single reference model without having to select or generate a different reference model for each test milk sample. Thus, the milk analyzer of the present subject matter is simpler in construction, portable, and can be easily used by the user in the dairy farms.

[0018] The manner in which the milk analyzer is used for conducting quantitative analysis of milk components is explained in detail, with respect to Figures la to 3. While aspects of the described milk analyzer for conducting quantitative analysis of milk components may be implemented in any number of different computing systems, environments, and/or configurations, the embodiments are described in the context of the following system(s).

[0019] Fig. la, Fig. lb, and Fig. lc illustrate a milk analyzer 100 for conducting quantitative analysis of milk components and an internal view of the milk analyzer 100 showing various hardware components in the milk analyzer 100, in accordance with an embodiment of the present subject matter. In said embodiment, the milk analyzer 100 includes a fiber-optic probe 102 implemented in a diffuse reflectance mode, a display unit 104, a light source 106, a near-infrared (NIR) spectrometer 108, and a controller 110 communicatively coupled with each other. The light source 106 emits light having wavelength ranging between 600 nanometers (nm) and 1100 nm. Examples of the light source 106 include white light source, such as halogen, xenon, and mercury lamp. In one implementation, the light source 106 may be powered by a power supply 1 12. The power supply 112 can also provide power to the other hardware components of the milk analyzer 100.

[0020] The emitted light is then transported to a interior of a test milk sample in a sample holder 103, for which spectra and quantities of milk components are unknown, through the fiber- optic probe 102 such that the emitted light is incident on the interior of the test milk sample. The fiber-optic probe 102, being implemented in the diffused reflectance mode, receives a reflected light from the interior of the test milk sample and transmits the reflected light to the NIR spectrometer 108 for performing spectroscopic analysis of the reflected light. The NIR spectrometer 108 can implement any of known methods such as diffraction gratings for filtering different wavelengths from the reflected light. In one implementation, the NIR spectrometer 108 can include a monochromator (not shown) which is used as a filter. Examples of monochromator include dielectric bandpass filters and double monochromators. Further, the NIR spectrometer 108 can include a detector (not shown) for detecting the reflected light transported by the fiberoptic probe 102. The detectors can be based on silicon based photoconductive materials.

[0021] The NIR spectrometer 108 performs the spectroscopic analysis of the reflected light to determine reflectance spectra of the test milk sample. In one implementation, parameters of the NIR spectrometer 108 such as data acquisition interval and data acquisition time are set based on hardware components used in the NIR spectrometer 108 such that data can be collected at reasonably long time intervals without significant loss of information. In one example, the data acquisition interval and the data acquisition time are set to 4 nm and 2 seconds, respectively. The parameters thus set enable the NIR spectrometer 108 to perform sampling and analysis of the reflected light using lower cost hardware components. The parameters of the NIR spectrometer 108 can be set using the display unit 104 through the interactive user-interface. The reflectance spectra are then provided to the controller 110 for conducting quantitative analysis of milk components of the test milk sample based on a single reference model, as described in later sections below. The results of the quantitative analysis, namely quantities of the milk components or component parts of the test milk sample can then be displayed on the display unit 104 through the interactive user-interface. The display of quantities of milk components can be easily understood and analyzed by a user for, for example, modifying diet of cattle and determining a price of the milk sample. Also, the user-interface can include various menu bars for indicating various functions performed by the milk analyzer such as performing self- calibration and initializing analysis.

[0022] Further, after analyzing one test milk sample, another test milk sample can be analyzed in a manner as described above, thereby enabling easy analysis of different milk samples. For the purpose, the fiber-optic probe 102 is cleansed each time before conducting successive quantitative analysis. The fiber-optic probe 102, being implemented in the diffuse reflectance mode, includes a set of orifices around a, circumference of the tip of the fiber-optic probe 102. The set of orifices are positioned such that tip of fiber-optic probe 102 can be easily cleansed by either a solvent, such as water, or a dry vapor stream. Therefore, the fiber-optic probe 102 can be easily and quickly cleansed for conducting quantitative analysis. Also, the cleansing is more effective and so the possibility of errors in measurement due to the fiber-optic probe 102 being contaminated with previous samples is considerably reduced.

[0023] As mentioned previously, the milk analyzer 100 can be used for conducting quantitative analysis of milk components based on a single reference model. In an example, the analyzed milk components are fat and SNF. The single reference model is a regression model generated based on spectral analysis of a set of reference milk samples, for which spectra and quantities of milk components of the set of reference milk samples are known. The single reference model can include a plurality of equations such that each equation can be used for determining the quantity of a different milk component in a milk sample. For example, the single reference model can include a first equation for determining fat content, a second equation for determining SNF content and so on. The quantities of milk components of the set of reference milk samples can be measured using any of known methods. The known spectra can be measured using same method used for performing spectral analysis of the set of reference milk samples. The set of reference milk samples include cow milk, buffalo milk, mixed milk, and milk from other sources such as sheep, goat, and camel. The single reference model thus determined is usable for quantification of milk components in any milk sample irrespective of the animal source and thus, unlike conventional methods, separate reference models are not required for quantification of components in milk samples from different animal sources.

[0024] In one implementation, the milk analyzer 100 can be used for generating the single reference model. In such an implementation, for each reference milk sample, a light from the light source 106 in the wavelength range of 600 nm and 1100 nm is transmitted through the fiber-optic probe 102 into the reference milk sample. In one implementation, the reference milk sample to be analyzed is taken in a sample holder (not shown) and a tip of the fiber-optic probe 102 is immersed in the reference milk sample and the light source 106 is activated through the user-interface on the display unit so that the light from the light source 106 is incident in the bulk of the reference milk sample. Reflected light from the reference milk sample is then received by the fiber-optic probe 102. The received reflected light is then transported by the fiber-optic probe 102 to the NIR spectrometer 108 for determining reflectance spectra based on spectral analysis of the reflected light. The reflectance spectra are then log transformed as absorbance data and then provided to the controller 110 for further processing.

[0025] Fig. 2a illustrates an example absorbance spectra 200 obtained by the milk analyzer 100 for one of the reference milk samples to determine quantity of component part, namely fat, based on an absorbance data. The wavelength in nanometers (nm) is provided along the horizontal axis labeled 202. The absorption spectra measured in absorption units (AU) is provided along the vertical axis 204.

[0026] Further, the controller 1 10 then uses the single reference model based on the absorbance data from each of the reference milk samples. The single reference model is generated based on known regression techniques such as Partial Least Squares (PLS) regression. The single reference model includes a plurality of equations, each equation corresponding to a different component in the test milk samples. In one example, the single reference model may include two equations, one each for determining quantities of fat and SNF, in the test milk sample. In one implementation, the controller 1 10 can use multivariate methods generated by spectral analysis software such as Unscrambler® X and Pirouette® Version 2.0, for the single reference model. In one implementation, a numerical second derivative is obtained before the regression analysis is performed on the reflectance spectra. The generated single reference model is stored in a memory associated with the controller 110 of the milk analyzer 100 for enabling quantitative analysis of test milk samples, for which quantities of component parts of the test milk sample are unknown. The single reference model, so generated from a wide range of different kinds of reference milk samples, can be used for conducting quantitative analysis of different kinds of test milk samples, such cow milk, buffalo milk, and mixed milk rather than a specific kind of test milk sample. The single reference model, thus, can be used for conducting the quantitative analysis of the test milk sample without a user, at a diary farm, having to input a source of the test milk sample in the milk analyzer. Therefore, the analysis process at the diary farm is much simplified and errors that may be caused due to erroneous input of the source are eliminated. Further, in one implementation, the generated single reference model is optimized for each milk analyzer based on the hardware components used in the milk analyzer, during manufacturing of the milk analyzer.

[0027] Further, the generated single reference model can be validated to determine accuracy of the quantitative analysis conducted by the milk analyzer 100. In one implementation, the generated single reference model is validated by comparing determined quantities of milk components of the reference milk sample with actual quantities of the milk components of each of the reference milk samples. The actual quantities of the milk components of each of the reference milk samples can be determined using any known methods. Based on the comparison, the accuracy of the quantitative analysis can be determined. In one implementation, the accuracy, so determined, can be saved in the memory in the milk analyzer 100 to provide the user a range within which the quantified amounts would lie.

[0028] Fig. 2b illustrates a calibration graph 206 based on comparison of quantities determined using single reference model and actual quantities determined using known methods. In an example, the calibration graph 206 shows predicted quantities fat that is determined based on the single reference model and actual quantities of the milk component. As shown in Fig. 2b, the predicted quantities are provided along the vertical axis 208 labeled as 'Predicted'. The actual quantities are provided along the horizontal axis 210 labeled as 'Lab'. It can be inferred from the calibration graph 206 that the predicted quantities are very close to actual quantities and the single reference model has a high accuracy.

[0029] Fig. 2c illustrates a calibration graph 212 based on comparison of quantities determined using single reference model and actual quantities determined using known methods. In an example, the calibration graph 212 shows predicted quantities of SNF that is determined based on the single reference model and actual quantities of the milk component. , As shown in Fig. 2c, the predicted quantities are provided along the vertical axis 214 labeled as 'Predicted'. The actual quantities are provided along the horizontal axis 216 labeled as 'Lab'. It can be inferred from the calibration graph 212 that the predicted quantities are very close to actual quantities and the single reference model has a high accuracy. [0030] As mentioned previously, the milk analyzer 100 can be used for conducting the quantitative analysis of milk components of test milk samples, for which spectra and quantities of milk components of the test milk samples are unknown, based on the .generated single reference model, which is pre-stored in the milk analyzer 100. The quantitative analysis of milk components of test milk samples is conducted by a user. The user can be an operator at a dairy farm. In operation, a background spectrum can be scanned at the beginning of the quantitative analysis without using any milk sample. The background spectrum can be collected to measure noise on a detector in the NIR spectrometer 108 when the light source 106 is not activated. In one implementation, the interactive user-interface on the display unit 104 is used for providing a prompt to initialize the milk analyzer 100 by scanning the background spectrum..

[0031] After obtaining the background spectrum, white reference values are obtained to determine intensity of light incident on the test milk sample at all wavelengths of interest. . The white reference values are obtained using a white reference material such as Spectralon®, which has high diffuse reflectance of the order of 99%. In one implementation, the white reference material is provided to the user while purchasing the milk analyzer 100. In operation, the fiberoptic 102 can be used to direct light onto and obtain reflected light from the white reference material. The reflected light can be analyzed by the NIR spectrometer 108 to determine the white reference values. The determined white reference values can be stored in the memory associated with the controller 1 10.

[0032] After the determination of the white reference values and the background spectrum, the milk analyzer 100 can be used for conducting quantitative analysis of milk components of test milk samples, for which spectra and quantities of milk components are unknown. In one implementation, absorption data of the test milk sample are determined in the same manner as described above for the reference milk sample. Further, in such implementation, the absorption data determined over the entire spectra is used for determining quantities of milk components. Also, the determined white reference values and the background spectrum are used to obtain the absorption data of the test milk sample. In one implementation, the interactive user- interface on the display unit 104 can be used for initiating the analysis process. Based on the determined absorption data, the controller 110 determines quantities of milk components of the test milk sample by substituting parameters of absorption spectra in the single reference model pre-stored in the memory associated with the controller 1 10. In one implementation, the milk components of the test milk sample are fat and SNF.

[0033] Further, the determined quantities of milk components of the test milk sample can be displayed on the display unit 104 through the interactive user-interface. Based on the determined quantities, diet of cattle in the dairy farm can be analyzed and modified accordingly. Also, price of the test milk can be assessed based on the determination of the quantities of milk components. In one implementation, the determined quantities of milk components are stored in the memory associated with the controller 1 10 for future reference and analysis, and can be later transferred to another device via a device interface, such as a USB interface. Thus, the milk analyzer 100 determines the quantities of the milk components and displays the determined quantities which a user at a diary farm can easily interpreted, thereby making the milk analyzer 100 a user friendly analytical instrument.

[0034] After analyzing one test milk sample, another milk sample can be analyzed in a manner as described above. For the purpose, the fiber-optic probe 102 is cleansed before conducting successive quantitative analysis. The fiber-optic probe 102, being implemented in the diffuse reflectance mode, includes a set of orifices around a circumference of the tip of the fiberoptic probe 102. The set of orifices are positioned such that tip of fiber-optic probe 102 is cleansed by either a solvent or a dry vapor stream. Therefore, the fiber-optic probe 102 can be easily and quickly cleansed for conducting successive quantitative analysis.

[0035] Fig 3 illustrates a method 300 of conducting quantitative analysis of milk components of test milk samples, in accordance with the embodiments of the present subject matter. The order in which the method 300 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method 300, or alternative method. Additionally, individual blocks may be deleted from the method 300 without departing from the spirit and scope of the subject matter described herein. In an example, the method 300 may be implemented in a milk analyzer, such as the milk analyzer 100.

[0036] As discussed previously, a single reference model is determined based on a spectral analysis of set of reference milk samples, for which spectra and quantities of milk components of the reference milk samples are known. The set of reference milk samples include cow milk, buffalo milk, mixed milk, and can also be milk from other animal sources, such as goat, sheep and camel. The single reference model is stored in a memory of the milk analyzer 100 for conducting quantitative analysis of milk components of test milk samples, for which spectra and quantities of milk components of the test milk samples are unknown.

[0037] Referring to the method 300, at block 302 a light having a wavelength range of 600 nm and 1 100 nm is transmitted to an interior of the test milk sample, such that the light is incident on the interior of the test milk sample. For example, the light source 106 emits a light in the wavelength range between 600 nm and 1 100 nm. The emitted light is then transmitted to the interior of the test milk sample through the fiber-optic probe 102 such that the light is incident on the interior of the test milk sample. In one implementation, the tip of the fiber-optic probe 102 is immersed in the test milk sample contained in a sample holder for transmitting the light to the interior of the test milk sample.

[0038] At block 304, a reflected light from the interior of the test milk sample is received by the fiber-optic probe 102 and transported to the NIR spectrometer 108. In one implementation, the NIR spectrometer 108 receives the reflected light at approximately every 4 nm.

[0039] At block 306, a reflectance spectra of test milk sample is determined based on spectral analysis of the received reflected light. For example, the NIR spectrometer 108 performs spectral analysis of the reflected light and determines reflectance spectra of the test milk sample. In one implementation, parameters of the NIR spectrometer 108 such as data acquisition interval and data acquisition time are set to 4 nm and 2 seconds, respectively, through the display unit 104. The parameters set at these value enable the NIR spectrometer 108 to perform sampling and analysis of the receive light with much faster rate at every 4nm interval. The reflectance spectra are then log transformed as absorbance data and then provided to the controller 110 for further processing.

[0040] At block 308, quantities of milk components of the test milk sample are determined based on second order derivative of the reflectance spectra and a predetermined single reference model. For example, the controller 1 10 process the determined reflectance spectra of the test milk sample based on the predetermined single reference model and determines the quantities of the milk components of the test milk sample. In one implementation, the second order derivative is determined for the reflectance spectra prior to determining quantities of the milk components. Although embodiments for the milk analyzer have been described in the language specific to structural features, it is to be understood that the invention is not necessarily limited to the specific features described herein. Rather, the specific features are disclosed and explained in the context of a few embodiments of the milk analyzer.

[0041] The milk analyzer of the present subject matter is not restricted to the embodiments that are mentioned above in the description. Although the subject matter has been described with reference to the specific embodiments, this description is not meant to be construed in limiting sense. Various modifications of the disclosed embodiments, as well as alternate embodiments of the subject matter, will become apparent to person skilled in the art upon reference to the description of the subject matter. It is therefore contemplated that such modifications can be made without departing from the spirit or the scope of the present subject matter as defined.

Claims

1/ We claim:

1. A milk analyzer (100) for conducting quantitative analysis of milk components of at least one test milk sample, the milk analyzer (100) comprising: a fiber-optic probe (102) implemented in a diffuse reflectance mode, wherein the fiber- optic probe (102) is configured to transmit a light in a preselected wavelength range between 600 nanometers (nm) and 1100 nm into the at least one test milk sample and to receive a reflected light from the at least one test milk sample; a near-infrared (NIR) spectrometer (108) communicatively coupled with the fiber-optic probe (102), wherein the NIR spectrometer (108) is configured to perform spectral analysis of the reflected light received by the fiber-optic probe (102) to determine reflectance spectra of the at least one test milk sample; and a controller (1 10) communicatively coupled with the NIR spectrometer (108), wherein the controller (1 10) is configured to conduct quantitative analysis of milk components of the at least one test milk sample based on the determined reflectance spectra and a predetermined single reference model to determine quantities of the milk components of the at least one test milk sample, and wherein the predetermined single reference model is generated based on a spectral analysis of a set of reference milk samples having known quantities of milk components of the reference milk samples.

2. The milk analyzer (100) as claimed in claim 1, wherein the milk analyzer (100) comprises a light source (106), the light source (106) communicatively coupled with the fiberoptic probe (102), for emitting the light in the preselected wavelength range into the at least one test milk sample.

3. The milk analyzer (100) as claimed in claim 1, wherein the milk analyzer (100) comprises a display unit (104) having an interactive user-interface, communicatively coupled with the controller (1 10), for displaying the determined quantities of milk components of the at least one test milk sample.

4. The milk analyzer (100) as claimed in claim 4, wherein the display unit (104) having the interactive user-interface is communicatively coupled with the NIR spectrometer (108), for setting a plurality of parameters of the NIR spectrometer (108).

5. The milk analyzer (100) as claimed in claim 1 , wherein the milk analyzer (100) determines quantities of milk components of the at least one test milk sample and the reference milk samples comprising fat and solid non fat (SNF).

6. The milk analyzer (100) as claimed in claim 1 , wherein the reference milk samples include cow milk, buffalo milk, mixed milk, and milk from other animal sources.

7. A method for conducting quantitative analysis of milk components of at least one test milk sample, the method implemented by a milk analyzer (100), the method comprising:

performing spectral analysis, by a near-infrared (NIK) spectrometer (108) in the milk analyzer (100), of a reflected light from the at least one test milk sample to determine reflectance spectra of the at least one test milk sample; and conducting, by a controller (1 10) in the milk analyzer (100), quantitative analysis of milk components the at least one test milk sample based on the determined reflectance spectra and a predetermined single reference model to determine quantities of milk components of the at least one test milk sample, wherein the predetermined single reference model is generated based on a spectral analysis of a set of reference milk samples having known quantities of milk components of the reference milk samples.

8. The method as claimed in claim 7, the method comprising: transmitting, by a fiber-optic probe (102) in the milk analyzer (100), a light in a preselected wavelength range between 600 nanometers (nm) and 1 100 nm into the at least one test milk sample; ' receiving, by the fiber-optic probe (102) in the milk analyzer (100), the reflected light from the at least one test milk sample; and providing, by the fiber-optic probe (102) in the milk analyzer (100), the reflected light to the NIR spectrometer ( 108) in the milk analyzer ( 100).

9. The method as claimed in claim 7, wherein the fiber-optic probe (102) is implemented in a diffuse reflectance mode.

10. The method as claimed in claim 7, the method comprising displaying, by a displaying unit (104) in the milk analyzer (100), the determined quantities of milk components of the at least one test milk sample.

11. The method as claimed in claim 10, the method comprising setting, by the displaying unit (104) in the milk analyzer (100), a plurality of parameters of the NIR spectrometer (108) in the milk analyzer (100).

12. The method as claimed in claim 7, wherein the ^J determining quantities of milk components of the at least one test milk sample and the reference milk ..samples comprises determining quantities fat and solid non fat (SNF).

13. The method as claimed in claim 7, wherein the reference milk samples include cow milk, buffalo milk, mixed milk, and milk from other animal sources.