CN108399617A - A kind of detection method and device of animal health condition - Google Patents
A kind of detection method and device of animal health condition Download PDFInfo
- Publication number
- CN108399617A CN108399617A CN201810151804.XA CN201810151804A CN108399617A CN 108399617 A CN108399617 A CN 108399617A CN 201810151804 A CN201810151804 A CN 201810151804A CN 108399617 A CN108399617 A CN 108399617A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- value
- animal
- gray
- 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.)
- Granted
Links
- 241001465754 Metazoa Species 0.000 title claims abstract description 76
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 230000036541 health Effects 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 29
- 230000003862 health status Effects 0.000 claims abstract description 23
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 10
- 230000002708 enhancing effect Effects 0.000 claims abstract description 5
- 238000004590 computer program Methods 0.000 claims description 17
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 9
- 230000001052 transient effect Effects 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 abstract description 7
- 230000008569 process Effects 0.000 abstract description 4
- 238000001914 filtration Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000037237 body shape Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of detection method and device of animal health condition, by the normal color image, depth image and the thermal infrared images that acquire animal；By normal color image gray processing, and after enhancing contrast, third image is obtained；The gray value of each pixel in third image and depth image is adjusted to preset value or less, the 4th image and the 5th image are obtained respectively, thermal infrared images gray processing is obtained into the second image, after second image, the 4th image and the 5th image are merged, trained convolutional neural networks are input to, the health status of animal is exported；To in the detection process to animal health condition, on the basis of normal color image and depth image after gray processing do not influence embodiment of the thermal infrared images after gray processing to the shell temperature information of animal, above-mentioned three kinds of image co-registrations are got up, to make the image after fusion embody more rich phenotypic characteristic, the robustness detected to animal health condition is improved.
Description
Technical field
The present invention relates to phenotypic characteristic detection technique fields, more particularly, to a kind of detection side of animal health condition
Method and device.
Background technology
The phenotypic characteristic of animal can reflect that the health status of animal, the phenotypic characteristic of animal are mainly wrapped to a certain extent
Include the build and shell temperature of animal, wherein the build of animal can be obtained by way of optical imagery, the body surface temperature of animal
Degree can be obtained by way of thermal infrared.
In recent years, it is that leading Artificial Intelligence Development is rapid with machine learning, weight is all achieved in numerous research fields
Quantum jump.Image detecting technique is an important research hotspot and difficult point in artificial intelligence field.The bodily form and shell temperature
As the phenotypic characteristic of animal, the health status of animal, therefore the phenotypic characteristic detection of animal can be reflected in different aspect
It is an important research direction.
Currently, the detection of the health status to animal, is based only upon the detection to the build of animal or is based only upon to animal
Shell temperature detection, the build of animal and shell temperature are not yet combined to the inspection realized to the health status of animal
It surveys, simultaneously as the posture of animal is different, optical imagery or thermal infrared imaging are easy to be influenced by illumination or complex environment,
The accuracy of the judgement to animal health condition is influenced, therefore at present the robustness of the detection of animal health condition is needed to be carried
It is high.
Invention content
In order to overcome the above problem or solve the above problems at least partly, the present invention provides a kind of animal health condition
Detection method and device.
According to an aspect of the present invention, a kind of detection method of animal health condition is provided, including：Acquire the mark of animal
Quasi color image, depth image and thermal infrared images；By normal color image and thermal infrared images gray processing, first is obtained respectively
Image and the second image, and enhance the contrast of the first image, obtain third image；By the gray scale of each pixel in third image
Value be adjusted to preset value hereinafter, obtain the 4th image, the gray value of each pixel in depth image is adjusted to preset value hereinafter,
The 5th image is obtained, preset value is the maximum gradation value in the 6th image, and the 6th image is acquired by animal under health status
Thermal infrared images gray processing after image；Second image, the 4th image and the 5th image are merged, the 7th figure is obtained
Picture；7th image is input to trained convolutional neural networks, exports the health status of animal.
Wherein, enhance the contrast of the first image, obtain third image, including：First image is input to Gabor filtering
Device exports the 8th image of preset quantity grain direction；8th image of preset quantity grain direction is merged, is obtained
Obtain third image.
Wherein, the 8th image of preset quantity grain direction is merged, obtains third image, including：To own
The gray value of pixel in 8th image at same position is compared, and determines the maximum gradation value of the pixel at each position,
Using the maximum gradation value of the pixel at each position as the gray value of the pixel of corresponding position in third image.
Wherein, the gray value of each pixel in third image is adjusted to preset value hereinafter, obtaining the 4th image, including：
The first maximum gradation value for determining pixel in third image, using the ratio of preset value and the first maximum gradation value as the first ratio
Example；The gray value of each pixel in third image is pressed into the first proportional zoom, obtains the 4th image.
Wherein, the gray value of each pixel in depth image is adjusted to preset value hereinafter, obtaining the 5th image, including：
The second maximum gradation value for determining pixel in depth image, using the ratio of preset value and the second maximum gradation value as the second ratio
Example；The gray value of each pixel in depth image is pressed into the second proportional zoom, obtains the 5th image.
Wherein, the second image, the 4th image and the 5th image are merged, obtains the 7th image, including：By the second figure
The gray value of pixel in picture, the 4th image and the 5th image at same position is compared, and determines the pixel at each position
Maximum gradation value, using the maximum gradation value of the pixel at each position as the ash of the pixel of corresponding position in the 7th image
Angle value.
Wherein, the gray value of the pixel in the second image, the 4th image and the 5th image at same position is compared
Before, further include：It will be under the second image, the 4th image and the 5th Image Adjusting to identical coordinate system by affine transformation；It is logical
It crosses coordinate value and determines pixel in the second image, the 4th image and the 5th image at same position.
Another aspect of the present invention provides a kind of detection device of animal health condition, including：At least one processor；
And at least one processor being connect with processor communication, wherein：Memory is stored with the program that can be executed by processor and refers to
It enables, processor caller is instructed to execute above-mentioned method.
Another aspect of the present invention provides a kind of computer program product, and the computer program product is non-including being stored in
Computer program in transitory computer readable storage medium, the computer program include program instruction, when the program instruction quilt
When computer executes, computer is made to execute above-mentioned method.
Another aspect of the present invention provides a kind of non-transient computer readable storage medium, and the non-transient computer is readable
Storage medium stores computer program, which makes computer execute above-mentioned method.
The detection method and device of a kind of animal health condition provided by the invention, by the normal color figure for acquiring animal
Picture, depth image and thermal infrared images；By normal color image gray processing, and after enhancing contrast, third image is obtained；By
The gray value of each pixel is adjusted to preset value hereinafter, obtaining the 4th image and the 5th figure respectively in three images and depth image
Thermal infrared images gray processing is obtained the second image, the second image, the 4th image and the 5th image is merged, obtained by picture
7th image；7th image is input to trained convolutional neural networks, exports the health status of animal；To animal
In the detection process of health status, on the one hand using after gray processing normal color image and depth image embody the body of animal
Type information embodies the shell temperature information of animal using the thermal infrared images after gray processing；On the other hand, by gray value
Processing, normal color image and depth image after gray processing do not influence body of the thermal infrared images after gray processing to animal
On the basis of the embodiment of table temperature information, above-mentioned three kinds of image co-registrations are got up, it is richer to make the image after fusion embody
Rich phenotypic characteristic, improves the robustness detected to animal health condition.
Description of the drawings
It, below will be to embodiment or the prior art in order to illustrate more clearly of the present invention or technical solution in the prior art
Attached drawing needed in description is briefly described, it should be apparent that, the accompanying drawings in the following description is the one of the present invention
A little embodiments for those of ordinary skill in the art without creative efforts, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the flow chart according to the detection method of the animal health condition of the embodiment of the present invention.
Specific implementation mode
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached in the embodiment of the present invention
Figure, is clearly and completely described the technical solution in the present invention, it is clear that described embodiment is a part of the invention
Embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making wound
The every other embodiment obtained under the premise of the property made labour, shall fall within the protection scope of the present invention.
In one embodiment of the invention, with reference to figure 1, a kind of detection method of animal health condition is provided, including：
S11 acquires normal color image, depth image and the thermal infrared images of animal；S12, by normal color image and thermal infrared figure
As gray processing, the first image and the second image are obtained respectively, and enhance the contrast of the first image, obtain third image；S13,
The gray value of each pixel in third image is adjusted to preset value hereinafter, the 4th image is obtained, by each picture in depth image
The gray value of member is adjusted to preset value hereinafter, obtaining the 5th image, and preset value is the maximum gradation value in the 6th image, the 6th figure
Image after the thermal infrared images gray processing that picture is acquired by animal under health status；S14, by the second image, the 4th image
It is merged with the 5th image, obtains the 7th image；7th image is input to trained convolutional neural networks by S15, output
The health status of animal.
Specifically, coloured image is a kind of special circumstances of multispectral image, each pixel is by three bases in coloured image
Come what is indicated, component constitutes human vision institute energy between (0,255), by three primary colours for color red (R), green (G), blue (B) three components
Observe a variety of colors.Although coloured image colouring information is abundant, for machine learning language, overgenerous face
Color information will increase the difficulty of image recognition, and gray level image can express out the gradient information of any shade, gradient letter
Breath is critically important for identification object.In the present embodiment, by coloured image gray processing, the identification that can be reduced to pictorial information is difficult
Degree.In order to preferably embody the physical characteristic of animal, can also the coloured image after gray processing be subjected to contrast enhancement processing,
To obtain third image.
The gray value of each pixel of depth image can be used for characterizing distance of the certain point apart from video camera in scene, can table
Up to the three-dimensional information of scene, the physical characteristic of animal can be embodied to a certain extent；Thermal-induced imagery is with photoelectric technology
The infrared ray specific band signal of detection object heat radiation, converts the signal into the image differentiated for human vision, animal
Thermal-induced imagery gray processing after, the gray value of pixel reflects the shell temperature of animal.In the present embodiment, by infrared chart
As gray processing, the second image is obtained.
However, the thermal infrared images after gray processing is for the coloured image of gray processing or depth image, picture
The intensity value ranges of member are relatively small, if above three image directly merged, may destroy the shell temperature information of animal,
And those abnormal shell temperature information, such as it is apparently higher than the temperature information of animal normal body temperature, it is whether to judge animal
The most important information of health；Before fusion, need to handle the coloured image and depth image of gray processing, this implementation
In example, the gray value of each pixel in the coloured image (i.e. third image) of gray processing is adjusted to preset value hereinafter, acquisition the
The gray value of each pixel in depth image is adjusted to preset value hereinafter, obtaining the 5th image, wherein this is default by four images
Value is the maximum gradation value in the 6th image, the thermal infrared images gray processing that the 6th image is acquired by animal under health status
Image afterwards eliminates the coloured image of gray processing and the gray value of depth image to gray processing after above-mentioned processing
The influence of the important information of animal body surface temperature can be embodied in thermal infrared images.
Preferably, above coloured image, depth image and thermal infrared images can simultaneously and same angle acquisition, with reduce pair
The treating capacity of picture, and above-mentioned second image, the 4th image and the 5th image are merged, obtain the 7th image；By the 7th
Image is input to trained convolutional neural networks, exports the health status of animal.Wherein, the specific implementation of convolutional neural networks
Mode is as follows：First layer convolutional layer is 11 × 11 using 96 convolution kernel sizes, and the convolution filter that step-length is 4 is filtered,
Then result is sent into maximum pond layer, maximum pond layer sets pond window as 3 × 3, step-length 2；Second layer convolutional layer makes
It is 5 × 5 with 256 convolution kernel sizes, the convolution filter that step-length is 1 is filtered, and result is then sent into maximum pond layer,
Maximum pond layer sets pond window as 3 × 3, step-length 2；Third layer convolutional layer is 3 × 3 using 384 convolution kernel sizes, step
A length of 1 convolution filter is filtered, and result is then sent into maximum pond layer, and maximum pond layer sets pond window as 3
× 3, step-length 2 finally exports the result of the health status of animal.
Normal color image, depth image and the thermal infrared images that the present embodiment passes through acquisition animal；By normal color figure
As gray processing, and after enhancing contrast, third image is obtained；By the gray value tune of each pixel in third image and depth image
It is whole that thermal infrared images gray processing is obtained into the second image to preset value hereinafter, obtain the 4th image and the 5th image respectively, by the
Two images, the 4th image and the 5th image are merged, and the 7th image is obtained；7th image is input to trained convolution god
Through network, the health status of animal is exported；Thus in the detection process to animal health condition, after on the one hand utilizing gray processing
Normal color image and depth image embody the body-shape information of animal, embodied using the thermal infrared images after gray processing dynamic
The shell temperature information of object；On the other hand, by the processing to gray value, the normal color image after gray processing and depth map
On the basis of not influencing the thermal infrared images after gray processing to the embodiment of the shell temperature information of animal, by above-mentioned three kinds of images
Fusion is got up, and to making the image after fusion embody more rich phenotypic characteristic, is improved and is detected to animal health condition
Robustness.
Based on above example, enhance the contrast of the first image, obtains third image, including：First image is inputted
To Gabor filter, the 8th image of preset quantity grain direction is exported；By the 8th image of preset quantity grain direction
It is merged, obtains third image.
Specifically, in two-dimensional space, if being superimposed with a Gaussian function using a trigonometric function (such as SIN function)
A Gabor filter is obtained, Gabor filter can extract space part frequency feature, be a kind of effective texture inspection
Survey tool, wherein the direction of texture in filtered image can also be determined by relevant parameter in setting Gaussian function.
The information in coloured image in order to adequately extract gray processing filters to obtain multiple grain directions by Gabor filter
Filtering image (i.e. the 8th image) then carries out the filtering image for filtering to obtain multiple grain directions by Gabor filter
Fusion.Preferably, grain direction is uniformly distributed on 0 °~180 ° directions, such as the quantity of grain direction selects 8, then corresponds to
The angle of grain direction be respectively 0 °, 22.5 °, 45 °, 67.5 °, 90 °, 112.5 °, 135 ° and 157.5 °.Wherein, Gabor is filtered
The expression formula of wave device is as follows：
Wherein, in formula, σ represents the scale parameter of Gabor filter, θkIndicate the angle of k-th of grain direction,For
The scale parameter in the directions x of Gabor filter,For the scale parameter in the directions y of Gabor filter,For k-th of texture
The x values of the angle in direction,For the y values of the angle of k-th of grain direction, fkFor the centre frequency of k-th of grain direction, n is
The total quantity of grain direction.
Based on above example, the 8th image of preset quantity grain direction is merged, obtains third image, packet
It includes：The gray value of pixel in all 8th images at same position is compared, determines pixel at each position most
High-gray level value, using the maximum gradation value of the pixel at each position as the gray scale of the pixel of corresponding position in third image
Value.
Specifically, filter to obtain the filtering image (i.e. the 8th image) of multiple grain directions by Gabor filter, in order to
Better image contrast is obtained, multiple 8th images are merged, the mode of fusion is will be identical in all 8th images
The gray value of pixel at position is compared, and the maximum gradation value of the pixel at each position is determined, at each position
Pixel is assigned a value of maximum gradation value, finally obtains blending image (i.e. third image), which has reached enhancing image pair
Than the purpose of degree.
Based on above example, the gray value of each pixel in third image is adjusted to preset value hereinafter, obtaining the 4th
Image, including：The first maximum gradation value for determining pixel in third image makees the ratio of preset value and the first maximum gradation value
For the first ratio；The gray value of each pixel in third image is pressed into the first proportional zoom, obtains the 4th image.
Specifically, in order to which the gray value of each pixel in third image is adjusted to preset value hereinafter, and ensureing third figure
The gray value of each pixel in third image, can in proportion be adjusted by the gradient information of the gray value as in, it is first determined
Maximum gradation value (i.e. the first maximum gradation value) in the gray value of all pixels of third image, will be default in above-described embodiment
Adjustment ratio (i.e. first ratio) of the ratio of value and the first maximum gradation value as the gray value of each pixel in third image,
Then the gray value of each pixel in third image is pressed into the first proportional zoom, obtains the 4th image.By above-mentioned processing procedure,
Remain the gradient information of the gray value in third image, it is thus also avoided that the thermal infrared of gray value in third image to gray processing
The influence of image.Wherein, the expression formula for scaling formula is as follows：
Wherein, I'a(x, y) is the gray value of pixel in the 4th image, Ia(x, y) is the gray value of pixel in third image,
C is preset value, max Ia(x, y) is the first maximum gradation value.
Based on above example, the gray value of each pixel in depth image is adjusted to preset value hereinafter, obtaining the 5th
Image, including：The second maximum gradation value for determining pixel in depth image makees the ratio of preset value and the second maximum gradation value
For the second ratio；The gray value of each pixel in depth image is pressed into the second proportional zoom, obtains the 5th image.
Specifically, in order to which the gray value of each pixel in depth image is adjusted to preset value hereinafter, and ensureing depth map
The gray value of each pixel in depth image, can in proportion be adjusted by the gradient information of the gray value as in, it is first determined
Maximum gradation value (i.e. the second maximum gradation value) in the gray value of all pixels of depth image, will be default in above-described embodiment
Adjustment ratio (i.e. second ratio) of the ratio of value and the second maximum gradation value as the gray value of each pixel in depth image,
Then the gray value of each pixel in depth image is pressed into the second proportional zoom, obtains the 5th image.By above-mentioned processing procedure,
Remain the gradient information of the gray value in depth image, it is thus also avoided that the thermal infrared of gray value in depth image to gray processing
The influence of image.Wherein, the expression formula for scaling formula is as follows：
Wherein, I 'b(x, y) is the gray value of pixel in the 5th image, Ib(x, y) is the gray value of pixel in depth image,
C is preset value, max Ib(x, y) is the second maximum gradation value.
Based on above example, the second image, the 4th image and the 5th image are merged, obtain the 7th image, packet
It includes：The gray value of pixel in second image, the 4th image and the 5th image at same position is compared, determines each
The maximum gradation value for setting the pixel at place, using the maximum gradation value of the pixel at each position as corresponding position in the 7th image
Pixel gray value.
Specifically, the process that the second image, the 4th image and the 5th image are merged with to multiple 8th images
Fusion process is similar, and details are not described herein.
Based on above example, by the gray scale of the pixel in the second image, the 4th image and the 5th image at same position
Before value is compared, further include：By affine transformation by the second image, the 4th image and the 5th Image Adjusting to identical seat
Under mark system；The pixel in the second image, the 4th image and the 5th image at same position is determined by coordinate value.
Specifically, for the ease of the position of pixel in determining second image, the 4th image and the 5th image, it can be by second
Under image, the 4th image and the 5th Image Adjusting to identical coordinate system, on the basis of one of image, affine transformation is utilized
Position adjustment is carried out to other two image, such as under rectangular coordinate system, the formula of affine transformation is as follows：
Wherein, x1For the x coordinate value of the image before affine transformation, y1For the y-coordinate value of the image before affine transformation, x2For
The x coordinate value of image after affine transformation, y2For the y-coordinate value of the image after affine transformation, txFor shift value along the x-axis direction,
tyFor shift value along the y-axis direction, s is zoom scale, and θ is the angle rotated counterclockwise using origin as axle center.
Changed by above-mentioned radiation, by under the second image, the 4th image and the 5th Image Adjusting to identical coordinate system, side
It will pass through coordinate value and determine pixel at same position in each image.
As another embodiment of the present invention, a kind of detection device of animal health condition is provided, including：At least one place
Manage device；And at least one processor being connect with processor communication, wherein：Memory is stored with the journey that can be executed by processor
Sequence instructs, the instruction of processor caller to execute the method that above-mentioned each method embodiment is provided, such as including：Acquire animal
Normal color image, depth image and thermal infrared images；By normal color image and thermal infrared images gray processing, obtain respectively
First image and the second image, and enhance the contrast of the first image, obtain third image；By each pixel in third image
Gray value is adjusted to preset value hereinafter, obtaining the 4th image, and the gray value of each pixel in depth image is adjusted to preset value
Hereinafter, obtaining the 5th image, preset value is the maximum gradation value in the 6th image, the 6th image institute under health status for animal
Image after the thermal infrared images gray processing of acquisition；Second image, the 4th image and the 5th image are merged, obtain the 7th
Image；7th image is input to trained convolutional neural networks, exports the health status of animal.
As another embodiment of the present invention, a kind of computer program product is provided, which includes
The computer program being stored in non-transient computer readable storage medium, the computer program include program instruction, work as program
Instruction is when being computer-executed, and computer is able to carry out the method that above-mentioned each method embodiment is provided, such as including：Acquisition is dynamic
Normal color image, depth image and the thermal infrared images of object；By normal color image and thermal infrared images gray processing, obtain respectively
The first image and the second image are obtained, and enhances the contrast of the first image, obtains third image；By each pixel in third image
Gray value be adjusted to preset value hereinafter, obtain the 4th image, the gray value of each pixel in depth image is adjusted to default
For value hereinafter, obtaining the 5th image, preset value is the maximum gradation value in the 6th image, and the 6th image is animal under health status
Image after the thermal infrared images gray processing acquired；Second image, the 4th image and the 5th image are merged, obtain the
Seven images；7th image is input to trained convolutional neural networks, exports the health status of animal.
As another embodiment of the present invention, a kind of non-transient computer readable storage medium is provided, the non-transient meter
Calculation machine readable storage medium storing program for executing stores computer program, which makes the above-mentioned each method embodiment of computer execution be carried
The method of confession, such as including：Acquire normal color image, depth image and the thermal infrared images of animal；By normal color image
With thermal infrared images gray processing, the first image and the second image are obtained respectively, and enhance the contrast of the first image, obtain third
Image；The gray value of each pixel in third image is adjusted to preset value hereinafter, obtaining the 4th image, it will be every in depth image
The gray value of one pixel is adjusted to preset value hereinafter, obtaining the 5th image, and preset value is the maximum gradation value in the 6th image, the
Image after the thermal infrared images gray processing that six images are acquired by animal under health status；By the second image, the 4th image
It is merged with the 5th image, obtains the 7th image；7th image is input to trained convolutional neural networks, exports animal
Health status.
One of ordinary skill in the art will appreciate that：Realize that all or part of step of above method embodiment can pass through
Computer program instructions relevant hardware is completed, and computer program above-mentioned can be stored in a computer-readable storage and be situated between
In matter, which when being executed, executes step including the steps of the foregoing method embodiments；And storage medium above-mentioned includes：
The various media that can store program code such as ROM, RAM, magnetic disc or CD.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be expressed in the form of software products in other words, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
What is finally illustrated is：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although ginseng
According to previous embodiment, invention is explained in detail, it will be understood by those of ordinary skill in the art that：It still can be with
Technical scheme described in the above embodiments is modified or equivalent replacement of some of the technical features；And
These modifications or replacements, the spirit and model of various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution
It encloses.
Claims (10)
1. a kind of detection method of animal health condition, which is characterized in that including：
Acquire normal color image, depth image and the thermal infrared images of animal；
By the normal color image and the thermal infrared images gray processing, the first image and the second image are obtained respectively, and increase
The contrast of strong described first image, obtains third image；
The gray value of each pixel in the third image is adjusted to preset value hereinafter, obtaining the 4th image；By the depth
The gray value of each pixel is adjusted to the preset value hereinafter, obtaining the 5th image in image；The preset value is the 6th image
In maximum gradation value, after the thermal infrared images gray processing that the 6th image is acquired by the animal under health status
Image；
Second image, the 4th image and the 5th image are merged, the 7th image is obtained；
7th image is input to trained convolutional neural networks, exports the health status of the animal.
2. according to the method described in claim 1, it is characterized in that, the contrast of the enhancing described first image, obtains the
Three images, including：
Described first image is input to Gabor filter, exports the 8th image of preset quantity grain direction；
8th image of the preset quantity grain direction is merged, the third image is obtained.
3. according to the method described in claim 2, it is characterized in that, the 8th figure by the preset quantity grain direction
Picture is merged, and the third image is obtained, including：
The gray value of pixel in all 8th images at same position is compared, determines pixel at each position most
High-gray level value, using the maximum gradation value of the pixel at each position as the ash of the pixel of corresponding position in the third image
Angle value.
4. according to the method described in claim 1, it is characterized in that, the gray value by each pixel in the third image
Preset value is adjusted to hereinafter, obtaining the 4th image, including：
The first maximum gradation value for determining pixel in the third image, by the preset value and first maximum gradation value
Ratio is as the first ratio；
The gray value of each pixel in the third image is pressed into first proportional zoom, obtains the 4th image.
5. according to the method described in claim 1, it is characterized in that, the gray value by each pixel in the depth image
The preset value is adjusted to hereinafter, obtaining the 5th image, including：
The second maximum gradation value for determining pixel in the depth image, by the preset value and second maximum gradation value
Ratio is as the second ratio；
The gray value of each pixel in the depth image is pressed into second proportional zoom, obtains the 5th image.
6. according to the method described in claim 1, it is characterized in that, described by second image, the 4th image and institute
It states the 5th image to be merged, obtains the 7th image, including：
The gray value of pixel in second image, the 4th image and the 5th image at same position is compared
Compared with the maximum gradation value of the pixel at each position being determined, using the maximum gradation value of the pixel at each position as described
The gray value of the pixel of corresponding position in seven images.
7. according to the method described in claim 6, it is characterized in that, described by second image, the 4th image and institute
State the pixel in the 5th image at same position gray value be compared before, further include：
By affine transformation by second image, the 4th image and the 5th Image Adjusting to identical coordinate system
Under；
The pixel in second image, the 4th image and the 5th image at same position is determined by coordinate value.
8. a kind of detection device of animal health condition, which is characterized in that including：
At least one processor；And at least one processor being connect with the processor communication, wherein：
The memory is stored with the program instruction that can be executed by the processor, the processor call described program instruction with
Execute the method as described in claim 1 to 7 is any.
9. a kind of computer program product, which is characterized in that the computer program product includes being stored in non-transient computer
Computer program on readable storage medium storing program for executing, the computer program include program instruction, when described program is instructed by computer
When execution, the computer is made to execute the method as described in claim 1 to 7 is any.
10. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited
Computer program is stored up, the computer program makes the computer execute the method as described in claim 1 to 7 is any.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810151804.XA CN108399617B (en) | 2018-02-14 | 2018-02-14 | Method and device for detecting animal health condition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810151804.XA CN108399617B (en) | 2018-02-14 | 2018-02-14 | Method and device for detecting animal health condition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108399617A true CN108399617A (en) | 2018-08-14 |
| CN108399617B CN108399617B (en) | 2020-08-14 |
Family
ID=63096552
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810151804.XA Expired - Fee Related CN108399617B (en) | 2018-02-14 | 2018-02-14 | Method and device for detecting animal health condition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108399617B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110889847A (en) * | 2019-12-10 | 2020-03-17 | 苏州大学 | Nuclear radiation damage assessment system and method based on infrared imaging |
| CN111104833A (en) * | 2018-10-29 | 2020-05-05 | 北京三快在线科技有限公司 | Method and apparatus for in vivo examination, storage medium, and electronic device |
| CN112241935A (en) * | 2019-07-18 | 2021-01-19 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment and storage medium |
| CN117912116A (en) * | 2024-01-24 | 2024-04-19 | 北京林业大学 | Wild animal posture estimation method in real scene |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009028930A1 (en) * | 2007-08-27 | 2009-03-05 | Maasland N.V. | System for and method of managing a group of animals |
| CN104616284A (en) * | 2014-12-09 | 2015-05-13 | 中国科学院上海技术物理研究所 | Pixel-level alignment algorithm for color images to depth images of color depth camera |
| CN106472412A (en) * | 2016-10-10 | 2017-03-08 | 重庆科技学院 | Pet feeding method and system based on internet of things |
| CN206699119U (en) * | 2017-03-31 | 2017-12-05 | 北京农业信息技术研究中心 | A kind of cultivated animals monitoring system |
-
2018
- 2018-02-14 CN CN201810151804.XA patent/CN108399617B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009028930A1 (en) * | 2007-08-27 | 2009-03-05 | Maasland N.V. | System for and method of managing a group of animals |
| CN104616284A (en) * | 2014-12-09 | 2015-05-13 | 中国科学院上海技术物理研究所 | Pixel-level alignment algorithm for color images to depth images of color depth camera |
| CN106472412A (en) * | 2016-10-10 | 2017-03-08 | 重庆科技学院 | Pet feeding method and system based on internet of things |
| CN206699119U (en) * | 2017-03-31 | 2017-12-05 | 北京农业信息技术研究中心 | A kind of cultivated animals monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 沈明霞 等: "畜禽养殖个体信息监测技术研究进展", 《农业机械学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111104833A (en) * | 2018-10-29 | 2020-05-05 | 北京三快在线科技有限公司 | Method and apparatus for in vivo examination, storage medium, and electronic device |
| CN112241935A (en) * | 2019-07-18 | 2021-01-19 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment and storage medium |
| CN112241935B (en) * | 2019-07-18 | 2023-05-26 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment and storage medium |
| CN110889847A (en) * | 2019-12-10 | 2020-03-17 | 苏州大学 | Nuclear radiation damage assessment system and method based on infrared imaging |
| CN117912116A (en) * | 2024-01-24 | 2024-04-19 | 北京林业大学 | Wild animal posture estimation method in real scene |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108399617B (en) | 2020-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3948764B1 (en) | Method and apparatus for training neural network model for enhancing image detail | |
| CN106548467B (en) | The method and device of infrared image and visual image fusion | |
| CN103430551B (en) | Use the imaging system and its operating method of the lens unit with axial chromatic aberration | |
| CN107977940A (en) | background blurring processing method, device and equipment | |
| Zheng et al. | Qualitative and quantitative comparisons of multispectral night vision colorization techniques | |
| CN103517046B (en) | Method, apparatus and computer program product for capturing video content | |
| US20100208946A1 (en) | Specifying flesh area on image | |
| CN108419028A (en) | Image processing method, device, computer readable storage medium and electronic equipment | |
| CN106960428A (en) | Visible ray and infrared double-waveband image co-registration Enhancement Method | |
| US20070098260A1 (en) | Detecting and correcting peteye | |
| CN108053363A (en) | Background blurring processing method, device and equipment | |
| CN105103187A (en) | Multi-spectral imaging system for shadow detection and attenuation | |
| CN108399617A (en) | A kind of detection method and device of animal health condition | |
| Finlayson | Colour and illumination in computer vision | |
| CN108710910A (en) | A kind of target identification method and system based on convolutional neural networks | |
| CN109461186A (en) | Image processing method, device, computer readable storage medium and electronic equipment | |
| CN110268420A (en) | In the picture detect background objects on exotic computer implemented method, in the picture detect background objects on exotic equipment and computer program product | |
| CN116681636B (en) | Light infrared and visible light image fusion method based on convolutional neural network | |
| JPWO2020003991A1 (en) | Medical image learning equipment, methods and programs | |
| CN108616700A (en) | Image processing method and device, electronic equipment, computer readable storage medium | |
| CN105979233B (en) | Demosaicing methods, image processor and imaging sensor | |
| CN110192388A (en) | Image processing apparatus, digital camera, image processing program and recording medium | |
| CN109191398A (en) | Image processing method, device, computer readable storage medium and electronic equipment | |
| CN107680070B (en) | Hierarchical weight image fusion method based on original image content | |
| JP2008077460A (en) | Image evaluation device and its program |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200814 Termination date: 20210214 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |