CN109785342A - A kind of analysis method of the animal meat quality uniformity - Google Patents
A kind of analysis method of the animal meat quality uniformity Download PDFInfo
- Publication number
- CN109785342A CN109785342A CN201910036147.9A CN201910036147A CN109785342A CN 109785342 A CN109785342 A CN 109785342A CN 201910036147 A CN201910036147 A CN 201910036147A CN 109785342 A CN109785342 A CN 109785342A
- Authority
- CN
- China
- Prior art keywords
- image
- meat
- picture
- region
- meat quality
- 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.)
- Pending
Links
- 235000013372 meat Nutrition 0.000 title claims abstract description 62
- 241001465754 Metazoa Species 0.000 title claims abstract description 22
- 238000004458 analytical method Methods 0.000 title claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000003044 adaptive effect Effects 0.000 claims abstract description 6
- 239000000284 extract Substances 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims abstract description 6
- 238000003709 image segmentation Methods 0.000 claims description 5
- 238000000034 method Methods 0.000 abstract description 14
- 235000013305 food Nutrition 0.000 abstract description 8
- 238000001514 detection method Methods 0.000 abstract description 6
- 235000015278 beef Nutrition 0.000 description 14
- 235000015277 pork Nutrition 0.000 description 14
- 235000013330 chicken meat Nutrition 0.000 description 13
- 230000006870 function Effects 0.000 description 10
- 239000000203 mixture Substances 0.000 description 7
- 241000287828 Gallus gallus Species 0.000 description 6
- 241001584785 Anavitrinella pampinaria Species 0.000 description 4
- 238000013459 approach Methods 0.000 description 4
- 230000008447 perception Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 239000010440 gypsum Substances 0.000 description 2
- 229910052602 gypsum Inorganic materials 0.000 description 2
- 235000020997 lean meat Nutrition 0.000 description 2
- 235000013622 meat product Nutrition 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000003205 fragrance Substances 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 239000004579 marble Substances 0.000 description 1
- 230000000050 nutritive effect Effects 0.000 description 1
- 235000020991 processed meat Nutrition 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000036548 skin texture Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The present invention discloses a kind of analysis method of animal meat quality uniformity, extract animal meat quality photo, the color format image for cutting gained picture is extracted using MATLAB software, gained color image extracts red channel, green channel and blue channel, obtains the gray processing picture in three kinds of channels using gray processing function;The clearest gray processing picture of meat texture is therefrom chosen to deal with objects as follow-up study;And the gray scale character matrix of the image is obtained, binaryzation then is carried out with adaptive threshold, obtains binary image;N number of meat image-region in binary image is chosen to calculate
Description
Technical field
The present invention relates to animal meat quality performance rating fields, and in particular to a kind of analysis method of the animal meat quality uniformity.
Background technique
China is maximum animal meat quality producing country and country of consumption in the world.The quality and safety of animal meat quality has become
The core topic of public's general concern, food quality detection are the key that guarantee one ring of Safety of Food Quality.
Between in the past few decades, because of the high and palatable taste of the nutritive value of animal meat quality, need of the consumer to it
It asks and sharply increases.But animal meat quality easily rots and pollution.In meat products industry, in order to reduce the processing cost of meat products,
The product of high quality is provided also for lasting, it is necessary to implement food quality control program.Uniformity is that processed meat food stuff is mostly important
One of processing quality attribute, the subjective scores that evaluation is often provided from well-trained professional group member.People
It is believed that preferably texture uniformity all has good front for organoleptic qualities such as succulence, tenderness and fragrance
It influences.Given this closely related with Safety of Food Quality, the high-precision detection of texture uniformity is that animal meat quality industry is most attached most importance to
One of work wanted.
Summary of the invention
The present invention is directed to animal meat quality meat uniformity decision problem, proposes a kind of analysis side of animal meat quality uniformity
Method, for judging animal meat quality quality, compared to previous meat quality judgment method, advantages of the present invention is embodied in: simple side
Just;The accuracy in detection of animal meat quality food quality is improved, practicability is high.
A kind of analysis method of the animal meat quality uniformity, comprising the following steps:
(1) shoot animal meat quality photo, using image segmentation algorithm extract photo intermediate region picture, and with 1000dpi ×
The spatial resolution of 1000dpi cuts picture;
(2) the color format image of gained picture is cut using MATLAB mathematical software extraction step (1), gained color image mentions
Red channel, green channel and blue channel are taken, obtains the gray scale in three kinds of channels using common gray processing function rgb2gray
Change picture;Three kinds of gray processing pictures are examined, the wherein clearest gray processing picture of texture is chosen and is handled as follow-up study
Object obtains the gray scale character matrix (0-255) of the picture, then carries out binaryzation with adaptive threshold, obtains binaryzation
Image;N number of meat image-region in continuous or any selection binary image, carries out the calculating of pixel distributing homogeneity, i.e.,
Calculate the partial deviations function of each image-region,=choose image-region area/entire binary picture image planes
The quantity for the pixel put in the quantity for the pixel number point put in product-selection image-region/entire binary image, N >=1;
(3) it is based on star deviation approach, proposes to define the side mean absolute deviation (Mean Absolute Discrepancy, MAD)
Method is calculated using step (2)And formula, i≤N calculates MAD value, and MAD gets over
Small, meat is more uniform, in order to enable the module introduced can obtain directly in terms of mathematical measure and human perception uniformity
The explanation and understanding of sight, spy provide mixture homogeneity as follows (Uniformity Coefficient, UC),, UC is bigger, and meat is more uniform.
When the UC >=0.5, otherwise it is uneven that meat, which is uniform,.
The features of the present invention and effect are as follows:
(1) the method for the present invention is simple and convenient, and practicability is higher.
(2) compared to previous meat quality judgment method, the accuracy in detection of animal meat quality food quality is improved, it is practical
Property it is high.
Detailed description of the invention
Fig. 1 is pork RGB image, gray level image and the ROI image figure of the embodiment of the present invention 1;
Fig. 2 be the embodiment of the present invention 1 pork texture analysis of Uniformity figure (binary image (a), choose image-region area/
Percentage-partial deviations function of entire binary image area(b)).
Specific embodiment
Substantive content of the invention is further illustrated with example with reference to the accompanying drawing, but the contents of the present invention are not limited to
This.
Embodiment 1
A kind of analysis method of the pork quality uniformity, comprising the following steps:
(1) image preprocessing: for the objectivity of science for proving the method for the present invention, the pig of the open report such as Huang (2013) is chosen
Meat image data analyzes the quality of pork in this research as research object;It is extracted among long saddle using image segmentation algorithm
The picture in region, and the long saddle intermediate region of pork sample is cut out with the spatial resolution of 1000dpi × 1000dpi
It cuts;Pork gray level image and ROI image (i.e. area-of-interest) are obtained using weighted average method, as shown in Figure 1;
(2) feature extraction: utilizing the pork color image of MATLAB software extraction step (1), and resulting pork color image mentions
Red channel, green channel and blue channel are taken, obtains the gray scale in three kinds of channels using common gray processing function rgb2gray
Change picture;Three kinds of gray processing pictures are examined, analysis is known: the RGB image of pork illustrates the big of pork quality well
Fibrous gypsum line feature;Red channel cannot obviously show the difference of lean meat and texture, and green channel can be more bright with blue channel
The difference of lean meat and texture is shown aobviously, and in contrast, green channel becomes apparent, and therefore, chooses green channel as pig
The gray level image of meat RGB image;Gray processing processing is carried out to selected picture with common MATLAB tool, obtains gray scale
Character matrix (0-255) then carries out binaryzation with adaptive threshold, obtains binary image;Arbitrarily choose N number of meat figure
As region, N is that the number of picture abscissa pixel, using this origin as angle point, takes two using the binary image lower left corner as origin
Value image abscissa pixel is horizontal edge, and binary image ordinate pixel is longitudinal edge, does rectangle, carries out the rectangular area
The calculating of interior pixel distributing homogeneity calculates the partial deviations function of each image-region,=choose figure
As the pixel number point put in region area/entire meat image area-selection image-region quantity/entire meat image in point
Pixel quantity, continuously take region, the region before selected areas includes later, until the entire two-value of region overlay taken
Change image, wherein region area is calculated as choosing the areal calculation formula of rectangle, it may be assumed that long multiplied by width;And the quantity of pixel is
The quantity (white is background) of black objects in binary image, such as Fig. 2 (a);
(3) uniformity calculates: being based on star deviation approach, proposes to define mean absolute deviation (Mean Absolute
Discrepancy, MAD) method, it is calculated using step (2)And formula, i≤
N calculates MAD value, and MAD is smaller, and meat is more uniform, in order to enable the module introduced can be in mathematical measure and human perception
Intuitive explanation and understanding are obtained in terms of uniformity, spy provides mixture homogeneity (Uniformity as follows
Coefficient, UC),, UC is bigger, and meat is more uniform;Choose the pig of the open report such as Huang (2013)
Meat image data extracts the textural characteristics in pork sample digital color image green channel ROI as research object, and green is logical
The skin texture detection result in road is as shown in Fig. 2, binary image (a), selection image-region area/entire binary image area
Percentage-partial deviations function(b);MAD value and UC value are calculated, the mixture homogeneity of pork sample green channel is UC
=0.8667, illustrate that distribution of the texture in the pork sample is that uniform namely pork sample meat quality is fine.
Embodiment 2
A kind of analysis method of the chicken meat uniformity, comprising the following steps:
(1) image preprocessing: the picture of chicken intermediate region is extracted using image segmentation algorithm, and to the middle area of chicken meat sample
Domain is cut with the spatial resolution of 1000dpi × 1000dpi;Using weighted average method obtain chicken gray level image with
And ROI image (i.e. area-of-interest);
(2) feature extraction: utilizing the chicken color image of MATLAB software extraction step (1), and resulting chicken color image mentions
Red channel, green channel and blue channel are taken, obtains and obtains the ash in three kinds of channels using common gray processing function rgb2gray
Degreeization picture;Three kinds of gray processing pictures are examined, analysis is known: the RGB image of chicken illustrates chicken meat well
Marble grain feature;Red channel cannot obviously show the difference of meat and texture, and green channel can be more bright with blue channel
The difference of meat and texture is shown aobviously, and in contrast, green channel becomes apparent, and therefore, chooses green channel as chicken
The gray level image of RGB image;Gray processing processing is carried out to selected picture with common MATLAB tool, obtains grey
Word matrix (0-255) then carries out binaryzation with adaptive threshold, obtains binary image;Arbitrarily choose 4 meat images
Region, 4 regions are located on four angles of binary image, i.e., do square respectively as vertex using four angles of binary image
Shape, 20 times of the size of a length of picture abscissa pixel of rectangle, width is 20 times of the size of picture ordinate pixel, into
The calculating of pixel distributing homogeneity in row rectangular area calculates the partial deviations function of each image-region,The quantity for the pixel number point put in=selection image-region area/entire meat image area-selection image-region/whole
The quantity for the pixel put in a meat image, wherein region area is calculated as choosing the areal calculation formula of rectangle, it may be assumed that length multiplies
With width;And in quantity, that is, binary image of pixel black objects quantity (white be background);
(3) uniformity calculates: being based on star deviation approach, proposes to define mean absolute deviation (Mean Absolute
Discrepancy, MAD) method, it is calculated using step (2)And formula, N=4,
MAD value is calculated, MAD is smaller, and meat is more uniform, in order to enable the module introduced can be equal with human perception in mathematical measure
Intuitive explanation and understanding are obtained in terms of even property, spy provides mixture homogeneity (Uniformity as follows
Coefficient, UC),, UC is bigger, and meat is more uniform;It is logical to extract chicken meat sample digital color image green
Textural characteristics in road ROI calculate MAD value and UC value, and the mixture homogeneity of chicken meat sample green channel is UC=0.5557, is said
Distribution of the bright texture in the chicken meat sample is that uniform namely chicken meat sample meat quality is fine.
Embodiment 3
A kind of analysis method of the beef meat uniformity, comprising the following steps:
(1) image preprocessing: the picture of beef intermediate region is extracted using image segmentation algorithm, and to the middle area of beef sample
Domain is cut with the spatial resolution of 1000dpi × 1000dpi;Using weighted average method obtain beef gray level image with
And ROI image (i.e. area-of-interest);
(2) feature extraction: utilizing the beef color image of MATLAB software extraction step (1), and resulting beef color image mentions
Red channel, green channel and blue channel are taken, obtains the gray scale in three kinds of channels using common gray processing function rgb2gray
Change picture;Three kinds of gray processing pictures are examined, analysis is known: the RGB image of beef illustrates the big of beef meat well
Fibrous gypsum line feature;Red channel cannot obviously show the difference of beef and texture, and green channel can be more bright with blue channel
The difference of beef and texture is shown aobviously, and in contrast, green channel becomes apparent, and therefore, chooses green channel as ox
The gray level image of meat RGB image;Gray processing processing is carried out to selected picture with common MATLAB tool, obtains gray scale
Character matrix (0-255) then carries out binaryzation with adaptive threshold, obtains binary image;Arbitrarily choose 1 meat figure
As region, which is located at the center of binary image, i.e., does rectangle, rectangle by dot of the center of binary image
200 times of size of a length of picture abscissa pixel, width is 200 times of the size of picture ordinate pixel, carries out square
The calculating of pixel distributing homogeneity in shape region calculates the partial deviations function of each image-region,=
Choose the quantity/entire meat figure for the pixel number point put in image-region area/entire meat image area-selection image-region
As the quantity of the pixel of interior point, wherein region area is calculated as choosing the areal calculation formula of rectangle, it may be assumed that long multiplied by width;And
The quantity of black objects in quantity, that is, binary image of pixel (white is background);
(3) uniformity calculates: being based on star deviation approach, proposes to define mean absolute deviation (Mean Absolute
Discrepancy, MAD) method, it is calculated using step (2)And formula, N=1,
MAD value is calculated, MAD is smaller, and meat is more uniform, in order to enable the module introduced can be equal with human perception in mathematical measure
Intuitive explanation and understanding are obtained in terms of even property, spy provides mixture homogeneity (Uniformity as follows
Coefficient, UC),, UC is bigger, and meat is more uniform;It is logical to extract beef sample digital color image green
Textural characteristics in road ROI calculate MAD value and UC value, mixture homogeneity UC=0.4885 of beef sample green channel, explanation
The meat quality of texture being unevenly distributed in the beef sample namely beef sample is general.
Claims (2)
1. a kind of analysis method of the animal meat quality uniformity, which comprises the following steps:
(1) shoot animal meat quality photo, using image segmentation algorithm extract photo intermediate region picture, and with 1000dpi ×
The spatial resolution of 1000dpi cuts picture;
(2) the color format image of gained picture is cut using MATLAB software extraction step (1), gained color image extracts red
Chrominance channel, green channel and blue channel obtain the gray processing picture in three kinds of channels using gray processing function;Choose wherein line
Clearest gray processing picture is managed, the gray scale character matrix of the picture is obtained, then binaryzation is carried out with adaptive threshold, obtains
Obtain binary image;N number of meat image-region in binary image is chosen, the partial deviations function of each image-region is calculated,=choose the quantity of pixel in image-region area/entire binary image area-selection image-region/
The quantity of pixel, N >=1 in entire binary image;
(3) it is calculated using step (2)And formula, MAD value is calculated, MAD is smaller,
Meat is more uniform,, UC is bigger, and meat is more uniform.
2. the analysis method of the animal meat quality uniformity according to claim 1, which is characterized in that when the UC >=0.5, meat
It is uniform, is otherwise uneven.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910036147.9A CN109785342A (en) | 2019-01-15 | 2019-01-15 | A kind of analysis method of the animal meat quality uniformity |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910036147.9A CN109785342A (en) | 2019-01-15 | 2019-01-15 | A kind of analysis method of the animal meat quality uniformity |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109785342A true CN109785342A (en) | 2019-05-21 |
Family
ID=66499414
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910036147.9A Pending CN109785342A (en) | 2019-01-15 | 2019-01-15 | A kind of analysis method of the animal meat quality uniformity |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109785342A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112034093A (en) * | 2019-12-05 | 2020-12-04 | 陈庆梅 | Planting uniformity field detection system and method |
CN112669268A (en) * | 2020-12-21 | 2021-04-16 | 昆明理工大学 | Method, system and terminal for evaluating distribution uniformity of multiphase mixed concentration field |
CN113281310A (en) * | 2021-04-06 | 2021-08-20 | 安徽工程大学 | Method for detecting light transmittance and uniformity of optical medium material |
CN116912887A (en) * | 2023-09-05 | 2023-10-20 | 广东省农业科学院动物科学研究所 | Broiler chicken breeding management method and system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706445A (en) * | 2009-11-10 | 2010-05-12 | 吉林大学 | Beef marbling grade scoring method and device |
CN103729642A (en) * | 2013-12-24 | 2014-04-16 | 中国科学院深圳先进技术研究院 | Meat quality detection system and method based on ultrasonic signals |
JP2014071018A (en) * | 2012-09-28 | 2014-04-21 | Obihiro Univ Of Agriculture & Veterinary Medicine | Evaluation method for marbling of meat |
CN104794502A (en) * | 2015-05-15 | 2015-07-22 | 哈尔滨工业大学 | Image processing and mode recognition technology-based rice blast spore microscopic image recognition method |
CN105067621A (en) * | 2015-08-17 | 2015-11-18 | 云南财经大学 | Method for judging multi-phase mixing uniformity |
-
2019
- 2019-01-15 CN CN201910036147.9A patent/CN109785342A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706445A (en) * | 2009-11-10 | 2010-05-12 | 吉林大学 | Beef marbling grade scoring method and device |
JP2014071018A (en) * | 2012-09-28 | 2014-04-21 | Obihiro Univ Of Agriculture & Veterinary Medicine | Evaluation method for marbling of meat |
CN103729642A (en) * | 2013-12-24 | 2014-04-16 | 中国科学院深圳先进技术研究院 | Meat quality detection system and method based on ultrasonic signals |
CN104794502A (en) * | 2015-05-15 | 2015-07-22 | 哈尔滨工业大学 | Image processing and mode recognition technology-based rice blast spore microscopic image recognition method |
CN105067621A (en) * | 2015-08-17 | 2015-11-18 | 云南财经大学 | Method for judging multi-phase mixing uniformity |
Non-Patent Citations (2)
Title |
---|
宋兰霞等: "基于计算机视觉的肉质大理石花纹含量的测定", 《云南农业大学学报(自然科学版)》 * |
张丽等: "边缘检测、二值化处理进行牛肉分级的应用", 《肉类研究》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112034093A (en) * | 2019-12-05 | 2020-12-04 | 陈庆梅 | Planting uniformity field detection system and method |
CN112669268A (en) * | 2020-12-21 | 2021-04-16 | 昆明理工大学 | Method, system and terminal for evaluating distribution uniformity of multiphase mixed concentration field |
CN113281310A (en) * | 2021-04-06 | 2021-08-20 | 安徽工程大学 | Method for detecting light transmittance and uniformity of optical medium material |
CN116912887A (en) * | 2023-09-05 | 2023-10-20 | 广东省农业科学院动物科学研究所 | Broiler chicken breeding management method and system |
CN116912887B (en) * | 2023-09-05 | 2023-12-15 | 广东省农业科学院动物科学研究所 | Broiler chicken breeding management method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109785342A (en) | A kind of analysis method of the animal meat quality uniformity | |
Veluchamy et al. | Image contrast and color enhancement using adaptive gamma correction and histogram equalization | |
Sun | Inspecting pizza topping percentage and distribution by a computer vision method | |
DE60009644T2 (en) | Color conversion using local information | |
Benedetti et al. | Color to gray conversions in the context of stereo matching algorithms: An analysis and comparison of current methods and an ad-hoc theoretically-motivated technique for image matching | |
Berezhnoy et al. | Computer analysis of van Gogh’s complementary colours | |
CN107507144B (en) | Skin color enhancement processing method and device and image processing device | |
El Khoury et al. | Color and sharpness assessment of single image dehazing | |
US10455123B2 (en) | Method for increasing the saturation of an image, and corresponding device | |
CN106529543B (en) | A kind of dynamic calculates the method and its system of polychrome grade binaryzation adaptive threshold | |
Stien et al. | Rapid estimation of fat content in salmon fillets by colour image analysis | |
CN114170622B (en) | Koi evaluation device, method, program, and storage medium | |
Yamamoto et al. | Efficient improvement method for separation of reflection components based on an energy function | |
Guang-rong | Rice color inspection based on image processing technique | |
CN108550124B (en) | Illumination compensation and image enhancement method based on bionic spiral | |
CN109285205A (en) | A kind of face exchanges method, system, computer readable storage medium and equipment | |
Wang et al. | Crop disease leaf image segmentation method based on color features | |
CN114170624B (en) | Koi evaluation system, and device, method, program, and storage medium for implementing koi evaluation system | |
Pei et al. | Color images enhancement using weighted histogram separation | |
Haldankar et al. | Converting gray scale image to color image | |
JP5057988B2 (en) | Melange yarn image simulation apparatus, method and program thereof | |
KR20030091471A (en) | YCrCb color based human face location detection method | |
Zulfikar et al. | Statistical investigation of skin image for disease analyzing in rural area using Matlab | |
Zhou et al. | Saliency preserving decolorization | |
Ko et al. | Color embedding and recovery using wavelet packet transform with pseudorandomized saturation code |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190521 |