CN107403181B - Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image - Google Patents
Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image Download PDFInfo
- Publication number
- CN107403181B CN107403181B CN201710402585.3A CN201710402585A CN107403181B CN 107403181 B CN107403181 B CN 107403181B CN 201710402585 A CN201710402585 A CN 201710402585A CN 107403181 B CN107403181 B CN 107403181B
- Authority
- CN
- China
- Prior art keywords
- sausage
- meat
- fat
- lean meat
- hyperspectral
- 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.)
- Expired - Fee Related
Links
- 235000013580 sausages Nutrition 0.000 title claims abstract description 74
- 235000020997 lean meat Nutrition 0.000 title claims abstract description 48
- 235000013372 meat Nutrition 0.000 title claims abstract description 38
- 238000000926 separation method Methods 0.000 title claims abstract description 6
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 238000009826 distribution Methods 0.000 claims abstract description 7
- 230000005484 gravity Effects 0.000 claims description 11
- 238000003064 k means clustering Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 13
- 238000005259 measurement Methods 0.000 abstract description 4
- 238000011897 real-time detection Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 239000000126 substance Substances 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 2
- 239000003517 fume Substances 0.000 description 2
- 235000013622 meat product Nutrition 0.000 description 2
- 230000001953 sensory effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 235000015241 bacon Nutrition 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000002542 deteriorative effect Effects 0.000 description 1
- 239000000539 dimer Substances 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 235000021049 nutrient content Nutrition 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Pathology (AREA)
- Probability & Statistics with Applications (AREA)
- Analytical Chemistry (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a lean meat and fat meat distribution detection method based on a Guangdong style sausage hyperspectral image, and solves the problems that after the lean meat and the lean meat in the sausage are mixed, the lean meat and the fat meat cannot be simply and effectively distinguished, and the ratio of the lean meat and the fat meat in the sausage is difficult to calculate. The sausage detection is carried out through the hyperspectral image, sample information can be rapidly and effectively obtained, and real-time detection and judgment are achieved. Compared with the prior art, the invention has the beneficial effects that: the method realizes the self-adaptive separation of lean meat and fat meat based on the hyperspectral image of the Cantonese sausage, does not need to perform determination by a chemical method, greatly simplifies the operation steps, shortens the detection time, and avoids the consequences of inaccurate measurement result and the like caused by artificial subjective factors by the self-adaptive K-means segmentation.
Description
Technical Field
The invention relates to the technical field of sausage quality nondestructive testing, in particular to a sausage near-infrared hyperspectral image-based lean meat and fat meat distribution detection method.
Background
Sausage (Sausage), commonly known as Sausage, is a Chinese-characteristic meat product prepared by cutting and mincing meat serving as a raw material into small pieces, adding auxiliary materials, filling animal casings, fermenting, maturing and drying, and is the largest product of Chinese meat products.
The Cantonese style sausage is divided into a top grade, a first grade and a second grade according to a new sanitary standard for pickled bacon products from 10 months and 1 day in 2005. At present, the quality identification and classification of the Cantonese sausage mainly adopt an empirical sensory evaluation method as a main method, namely a sampling method is adopted, an experienced sausage technician directly observes the color and the form of the sausage body by naked eyes, samples are steamed for 10-15 minutes, and taste flavor is identified. The sausage quality detection and classification are a multi-index comprehensive evaluation process, and the shape, color, taste, internal nutrient content and other indexes of the sausage are integrated. Obviously, the traditional grading method mainly based on sensory evaluation cannot meet the requirement of sausage quality detection. The laboratory instrument analysis method has the characteristics of high detection precision, good reliability and the like, but has the disadvantages of low detection speed, complex operation and high detection cost, and is difficult to carry out real-time detection on a production field. The detection technology represented by the electronic nose, the spectrum and the machine vision has the advantages of rapidness, no damage, low cost and the like, and has good application prospect in the quality detection of agricultural products and foods.
The hyperspectral imaging technology integrates the advantages of the traditional imaging technology and the spectrum technology, and the acquired hyperspectral image has the characteristic of 'map integration', namely simultaneously contains image information and spectrum information. The image information can be used to detect external quality, while the spectral information can be used to detect internal quality and security. The hyperspectral image is an optical image at a series of light wave wavelengths, has higher spectral resolution than the multispectral image, usually can reach 2-3 nm, and the measurement spectral range can be in ultraviolet, visible light and near infrared regions. The hyperspectral data is a three-dimensional image block where the two dimensions represent image pixel information (in coordinates X and Y) and the third dimension is wavelength information (in coordinates X and Y)
Denoted), the image blocks of the sample obtained at n wavelengths by the image detector array having a resolution of X Y pixels are a three-dimensional array of X Y n.
Disclosure of Invention
The invention provides a method for detecting the distribution of lean meat and fat meat based on a sausage near-infrared hyperspectral image, and solves the problems that after the fat meat and the lean meat in the sausage are mixed, the fat meat and the lean meat cannot be simply and effectively distinguished, and the ratio of the lean meat and the fat meat in the sausage is difficult to calculate. The sausage detection is carried out through the hyperspectral image, sample information can be rapidly and effectively obtained, and real-time detection and judgment are achieved.
A lean meat and fat meat self-adaptive separation method based on a Cantonese sausage hyperspectral image is characterized by comprising the following steps:
(1) acquiring near-infrared band hyperspectral original information of a Cantonese sausage sample under a casing-wrapped lossless condition by using a hyperspectral meter;
(2) performing Gaussian smoothing pretreatment on the hyperspectral original information obtained in the step (1) to obtain a pretreatment mapping image, and performing threshold segmentation between the sausage and the brown background to obtain a preliminary sausage ROI (region of interest) region;
(3) taking each sample point of the ROI area containing the sausage sample in the step (2) as a judgment unit, taking the near-infrared hyperspectral value of each point as a characteristic, adopting a K-means clustering algorithm, selecting Euclidean distance as the distance, and classifying the clustering number into two categories, and establishing a lean meat and fat meat distinguishing model of the Cantonese style sausage based on the hyperspectral characteristics;
(4) dividing the whole sausage area into two parts according to the hyperspectral information through the processing in the step (2) and the step (3), wherein the part with a large number of divided sample points is a lean meat part of the sausage, and the part with a large number of divided sample points is a fat meat part, and respective clustering centers of the two types are obtained through a K-means algorithm. And determining the membership degree of each category according to the distance ratio between the sample point in the feature space and the centers of the two clusters.
(5) And (4) visually displaying the membership degree of each sample point obtained in the step (4) according to the original image to obtain a lean meat and fat meat distribution map of the original sausage.
Further, in the step (2), the near-infrared hyperspectral image of the sausage placed on the brown background is subjected to threshold segmentation through the maximum between-class variance to extract the sausage contour.
Further, in the step (3), a K-means algorithm is adopted to carry out dimer classification, and Euclidean distance is selected as a distance measurement standard.
Further, in the step (3), 256 bands of reflected light of the near infrared 874.11 nm-1730.52 nm are selected as the characteristic vector of the sample point.
Further, in the step (4), the coordinates of the center point of the euclidean distance in the sample space are calculated for each of the sample points belonging to the lean meat and the fat meat, and the calculated coordinates are used as the respective clustering centers of the lean meat and the fat meat.
Further, in the step (4), a gravity model is introduced, the square of the reciprocal of the distance from the sample point to the center of the lean meat is used as the gravity of the lean meat to the sample point, the gravity of the fat meat to the sample point is considered in the same way, the ratio of the gravity of the lean meat to the sum of the gravity is used as the membership degree of the lean meat, and the membership degree of the sample point to the fat meat is considered in the same way.
Further, in the step (5), according to the original hyperspectral mapping image, pseudo-color display membership degree information is carried out, and lean meat and fat meat distribution information can be conveniently and visually observed.
Compared with the prior art, the invention has the beneficial effects that: the method realizes the self-adaptive separation of lean meat and fat meat based on the hyperspectral image of the Cantonese sausage, does not need to perform determination by a chemical method, greatly simplifies the operation steps, shortens the detection time, and avoids the consequences of inaccurate measurement result and the like caused by artificial subjective factors by the self-adaptive K-means segmentation.
Drawings
Fig. 1 is a hyperspectral image of a sausage sample under near infrared.
Fig. 2 is a view of a sausage image after threshold segmentation, with the red portion being the selected sausage region.
FIG. 3 is a two-classification hyperspectral image after K-means clustering, with red part being lean meat and blue part being fat meat.
FIG. 4 is a comparison of an original near-infrared hyperspectral image and a fat and thin separated image of 5 samples.
Detailed Description
The invention is further explained below in connection with fig. 1-4.
In the invention, the Cantonese sausage is purchased from two brands of Huangshanghuang and Xishangxi, and comprises 81 samples of a second-grade sausage, a first-grade sausage and a special-grade sausage, and is frozen and stored, and unfrozen in a sterile fume hood for 4h to normal temperature before experiment.
Firstly, taking and unfreezing the sausage, and in order to prevent the sausage from deteriorating, the sausage is unfrozen by a fume hood without water vapor interference.
Secondly, 1 group of unfrozen lossless sausage samples are taken, and hyperspectral original information of the Cantonese sausage is obtained by adopting a hyperspectral imager V10E-QE (Specim, Finland). The hyperspectral imager adopts a near-infrared camera, the exposure time is 3ms, the distance from a lens to a sample is 31.2cm, the moving direction of the sample is consistent with the axial direction of the sausage, the moving speed of the sample along with a platform is 26.4mm/s, and the sample longitudinally sweeps the whole sausage. The whole experimental process was carried out under constant temperature (about 16 ℃). Fig. 1 shows a hyperspectral image of a sausage sample under near infrared.
And thirdly, opening the hyperspectral data after black and white board correction through software ENVI 5.2, removing noise and smoothing individual overexposure points through Gaussian low-pass filtering of a 3 x 3 template to obtain a preprocessed mapping image, performing threshold segmentation on the hyperspectral image obtained through preprocessing by adopting a maximum inter-class variance method, and separating the sausage from the background to obtain an ROI (region of interest) corresponding to the sausage. Fig. 2 is a view of a thresholded segmented sausage image in which the red portion is the selected sausage region. And fourthly, performing self-adaptive binary classification on the ROI obtained in the last step by adopting a K-means clustering algorithm, wherein the software is realized by adopting matlab 2015 b. The K-means algorithm is a self-adaptive clustering algorithm, points in each ROI are corresponding to a feature space according to hyperspectral features, initial points (2) are determined as clustering kernels through a centroid method, the nearest kernels are searched for as new classes of sample points in each cycle of sample points, then the sample points of each class are averaged to obtain new kernels, and iteration is performed until the positions of the kernels do not move, so that the classes of the sample points are obtained. With both cores as the center of the respective classes. Please refer to fig. 3, which is a two-classification hyperspectral image after K-means clustering, wherein the red part is lean meat and the blue part is fat meat.
And fifthly, as the lean meat and the fat in the sausage are often fused together, one part often contains both the lean meat and the fat, and the part of information can be lost by adopting the dichotomy. Thus, membership was introduced here to evaluate whether each sample point belongs to lean body fatAnd (4) proportion. Referring to the gravity model, 1/r2(r is the distance to a certain type of nucleus) as the sample point is subjected to the force of the nucleus, and therefore has
Wherein, FIG. 4 is the image comparison of 5 samples of original near-infrared hyperspectral images and fat and lean separated images.
And (3) carrying out weighted average on each sausage point in the whole sausage according to the membership degree of the lean meat to obtain hyperspectral information of the whole sausage corresponding to the lean meat.
Claims (6)
1. A lean meat and fat meat self-adaptive separation method based on a Cantonese sausage hyperspectral image is characterized by comprising the following steps:
(1) acquiring near-infrared band hyperspectral original information of a sausage sample under a casing-wrapped lossless condition by adopting a hyperspectral camera and a near-infrared spectrometer;
(2) performing Gaussian smoothing pretreatment on the hyperspectral original information obtained in the step (1) to obtain a pretreatment mapping image, and performing threshold segmentation between the sausage and the brown background to obtain a sausage ROI (region of interest);
(3) taking each sample point of the ROI area containing the sausage sample in the step (2) as a judgment unit, taking the near-infrared hyperspectral value of each point as a characteristic, adopting a K-means clustering algorithm, selecting Euclidean distance as the distance, and classifying the clustering number into two categories, and establishing a lean meat and fat meat distinguishing model of the Cantonese style sausage based on the hyperspectral characteristics;
(4) dividing the whole sausage area into two parts according to hyperspectral information through the processing of the step (2) and the step (3), wherein the one with a large number of divided sample points is a lean meat part of the sausage, otherwise a fat meat part, two types of respective clustering centers are obtained through a K-means algorithm, and the membership degrees to the respective types are determined according to the distance value ratio of the sample points to the two clustering centers in the feature space:
Llean meatThe membership degree of lean meat of the sample points is as follows:
LfatThe fat membership of the sample points was:
Rlean meatDistance of sample point to lean nucleus, RFatIs the distance from the sample point to the fat core;
(5) and (4) visually displaying the membership degree of each sample point obtained in the step (4) according to the original image to obtain a lean meat and fat meat distribution map of the original sausage.
2. The method for adaptively separating lean meat and fat meat based on the Cantonese style sausage hyperspectral image as claimed in claim 1, wherein in the step (2), the sausage contour is extracted by performing threshold segmentation on the near-infrared hyperspectral image of the sausage placed on the background through the maximum between-class variance.
3. The method for adaptively separating lean meat and fat meat based on the hyperspectral image of the Cantonese sausage as claimed in claim 1, wherein in the step (3), the reflected light of 256 wave bands of near infrared 874.11 nm-1730.52 nm is selected as the feature vector of the sample point.
4. The method for adaptively separating lean meat from fat meat based on the hyperspectral image of the Cantonese sausage as claimed in claim 1, wherein in the step (4), the coordinates of the center point of the Euclidean distance in the sample space are calculated for each sample point belonging to the lean meat and the fat meat, and are used as the clustering center of each of the lean meat and the fat meat.
5. The method for adaptively separating lean meat from fat meat based on the hyperspectral image of the Cantonese sausage as claimed in claim 1, wherein in the step (4), a gravity model is introduced, the square of the reciprocal of the distance from the sample point to the center of the lean meat is taken as the gravity of the lean meat to the sample point, the gravity of the fat meat to the sample point is taken as the principle, the ratio of the gravity of the lean meat to the sum of the gravity is taken as the membership degree of the lean meat, and the membership degree of the sample point to the fat meat is taken as the principle.
6. The method for adaptively separating lean meat and fat meat based on the hyperspectral image of the Cantonese sausage as claimed in claim 1, wherein in the step (5), the membership information is displayed in a pseudo-color manner according to the original hyperspectral mapping image, so that the information of the distribution of the lean meat and the fat meat can be conveniently and visually observed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710402585.3A CN107403181B (en) | 2017-06-01 | 2017-06-01 | Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710402585.3A CN107403181B (en) | 2017-06-01 | 2017-06-01 | Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107403181A CN107403181A (en) | 2017-11-28 |
| CN107403181B true CN107403181B (en) | 2020-05-12 |
Family
ID=60405058
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710402585.3A Expired - Fee Related CN107403181B (en) | 2017-06-01 | 2017-06-01 | Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107403181B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109406446A (en) * | 2018-10-12 | 2019-03-01 | 四川长虹电器股份有限公司 | To the preprocess method and its call method of near-infrared data |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150580A (en) * | 2013-03-18 | 2013-06-12 | 武汉大学 | Method and device for Hyperspectral image semi-supervised classification |
| CN103257118A (en) * | 2013-04-22 | 2013-08-21 | 华南理工大学 | Fish tenderness hyperspectral detection method based on characteristic wave band |
| CN103914705A (en) * | 2014-03-20 | 2014-07-09 | 西安电子科技大学 | Hyperspectral image classification and wave band selection method based on multi-target immune cloning |
| CN104655585A (en) * | 2014-09-29 | 2015-05-27 | 南京农业大学 | PSE meat screening method based on near infrared spectrum |
| WO2016120463A1 (en) * | 2015-01-31 | 2016-08-04 | Ventana Medical Systems, Inc. | Systems and methods for area-of-interest detection using slide thumbnail images |
| CN105866063A (en) * | 2016-04-08 | 2016-08-17 | 北京工商大学 | Detection method for quality grade of sausage |
| CN106198423A (en) * | 2016-09-12 | 2016-12-07 | 电子科技大学 | A kind of method differentiating ham sausage grade based on visible and near infrared spectrum analytical technology |
| CN106339674A (en) * | 2016-08-17 | 2017-01-18 | 中国地质大学（武汉） | Hyperspectral image classification method based on edge preservation and graph cut model |
-
2017
- 2017-06-01 CN CN201710402585.3A patent/CN107403181B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150580A (en) * | 2013-03-18 | 2013-06-12 | 武汉大学 | Method and device for Hyperspectral image semi-supervised classification |
| CN103257118A (en) * | 2013-04-22 | 2013-08-21 | 华南理工大学 | Fish tenderness hyperspectral detection method based on characteristic wave band |
| CN103914705A (en) * | 2014-03-20 | 2014-07-09 | 西安电子科技大学 | Hyperspectral image classification and wave band selection method based on multi-target immune cloning |
| CN104655585A (en) * | 2014-09-29 | 2015-05-27 | 南京农业大学 | PSE meat screening method based on near infrared spectrum |
| WO2016120463A1 (en) * | 2015-01-31 | 2016-08-04 | Ventana Medical Systems, Inc. | Systems and methods for area-of-interest detection using slide thumbnail images |
| CN105866063A (en) * | 2016-04-08 | 2016-08-17 | 北京工商大学 | Detection method for quality grade of sausage |
| CN106339674A (en) * | 2016-08-17 | 2017-01-18 | 中国地质大学（武汉） | Hyperspectral image classification method based on edge preservation and graph cut model |
| CN106198423A (en) * | 2016-09-12 | 2016-12-07 | 电子科技大学 | A kind of method differentiating ham sausage grade based on visible and near infrared spectrum analytical technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107403181A (en) | 2017-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bhargava et al. | Fruits and vegetables quality evaluation using computer vision: A review | |
| Mazen et al. | Ripeness classification of bananas using an artificial neural network | |
| Xiao-bo et al. | In-line detection of apple defects using three color cameras system | |
| Blasco et al. | Recognition and classification of external skin damage in citrus fruits using multispectral data and morphological features | |
| Barbin et al. | Near-infrared hyperspectral imaging for grading and classification of pork | |
| Xing et al. | Combination of chemometric tools and image processing for bruise detection on apples | |
| Gao et al. | Application of hyperspectral imaging technology to discriminate different geographical origins of Jatropha curcas L. seeds | |
| Li et al. | Detection of early decayed oranges based on multispectral principal component image combining both bi-dimensional empirical mode decomposition and watershed segmentation method | |
| Abdullah et al. | Automated inspection system for colour and shape grading of starfruit (Averrhoa carambola L.) using machine vision sensor | |
| CN114663821B (en) | Real-time nondestructive detection method for product quality based on video hyperspectral imaging technology | |
| CN103528967A (en) | Hyperspectral image based overripe Lonicera edulis fruit identification method | |
| Alfatni et al. | Recent methods and techniques of external grading systems for agricultural crops quality inspection-review | |
| CN110929787B (en) | Apple objective grading system based on image | |
| Payman et al. | Development of an expert vision-based system for inspecting rice quality indices | |
| Yro et al. | Cocoa beans fermentation degree assessment for quality control using machine vision and multiclass svm classifier | |
| Torres et al. | Setting up a methodology to distinguish between green oranges and leaves using hyperspectral imaging | |
| Wang et al. | Application of hybrid image features for fast and non-invasive classification of raisin | |
| CN112132792B (en) | Hyperspectrum and deep learning-based pixel-level detection method for endogenous foreign matters in pecans | |
| CN107403181B (en) | Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image | |
| Janardhana et al. | Computer aided inspection system for food products using machine vision—a review | |
| Shi et al. | Application of spectral features for separating homochromatic foreign matter from mixed congee | |
| Ji et al. | Apple color automatic grading method based on machine vision | |
| Saad et al. | Bio-inspired vision fusion for quality assessment of harumanis mangoes | |
| Carrillo et al. | Artificial vision to assure coffee-Excelso beans quality | |
| Liu et al. | Paddy variety identification using hyperspectral imagery under non-ideal illumination conditions |
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: 20200512 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |