CN105243390B - Insect image identification detection method and classification of insect method - Google Patents
Insect image identification detection method and classification of insect method Download PDFInfo
- Publication number
- CN105243390B CN105243390B CN201510638158.6A CN201510638158A CN105243390B CN 105243390 B CN105243390 B CN 105243390B CN 201510638158 A CN201510638158 A CN 201510638158A CN 105243390 B CN105243390 B CN 105243390B
- Authority
- CN
- China
- Prior art keywords
- image
- insect
- value
- classification
- detection method
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/758—Involving statistics of pixels or of feature values, e.g. histogram matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Library & Information Science (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of insect image identification detection method and classification of insect methods, include the following steps:Acquisition is converted to each frame image in video data to gray scale by colour, and is split after the gray level image is carried out difference operation with corresponding modeling background;The characteristic value for calculating each image block, by characteristic value normalization, fusion treatment;According to maximum variance between clusters, adaptive threshold fuzziness is carried out to the Saliency maps of the super-pixel precision of gained, obtains binary map;The iamge description of image is generated, and each image carries out images match by it and in knowledge data base, according to matched as a result, the Known Species whether trypetid to be sorted belong in knowledge data base judged.The present invention realizes the automatic identification of caste by computer pattern recognition, and vision significance is combined well with phrase, can more accurately reflect the salient region in query image, improve the accuracy of identification.
Description
Technical field
The present invention relates to classification of insect methods, and in particular to a kind of insect image identification detection method and classification of insect method.
Background technology
Science of heredity is the science of the heredity and variation of studying organism, belongs to an important branch of life science, perhaps
Other more branches, for example, genetic engineering, is the emerging science being derived to the research institute of gene based on science of heredity.Due to
Gene contains the biological secret of itself with heredity, can observe and explain the biological phenomenon of macroscopic, while base by microcosmic angle
Because the development of science of heredity may also solve some significant problems of human society since ancient times, such as aging and disease, therefore, respectively
There's no one who doesn't or isn't pay attention to the research to gene in boundary.
The gene dosage of drosophila about only has the one third of the mankind, however, the control of the gene and the mankind of control drosophila development
Gene processed is similar, and drosophila life cycle it is short, can genetic experiment is appropriate for the advantages of such as mass propagation so that drosophila is in base
Because playing the part of important role in research.Using drosophila as the material of genetics research, using mutant strain study gene and character it
Between relationship nearly 100 years, so far, the tools of various research science of heredity have reached perfect stage, and drosophila furnished us with to today
The knowledge of science of heredity have its indelible contribution.
In the gene research of drosophila, genetic engineering researcher needs to collect the female drosophila not mated, to ensure
It will not be led to the failure of an experiment by the pollution of other male genes in genetic experiment operation.It is intended to recognize the female drosophila not mated,
Traditional method is that drosophila is fixed under microscope bottom with manpower, and is selected by the abdomen feature of micro- sem observation drosophila
Go out the female drosophila not mated.But since drosophila is a mass propgation, and female drosophila sprout wings 8 hours after reach sexal maturity,
Therefore this to have to last for carrying out with same interlude in such a way that human eye differentiates the female drosophila not mated, to avoid female fruit
Fly mates in cultivating accumulator tank.The above-mentioned suitable labor intensive of the practice for selecting the female drosophila not mated, while with people
Eye judges that the mode efficiency of drosophila feature is bad, often discards and largely can not to precisely pick out the female drosophila not mated
Determine the drosophila of feature.
Other than drosophila, different types of insect or the feature different in type insect are whether differentiated, together
Sample is all to need again to be judged by the insect under eye-observation microscope bottom, therefore also all there are conventional sorting methods to expend
Manpower, time and the bad problem of efficiency.
Invention content
To solve the above problems, the present invention provides a kind of insect image identification detection method and classification of insect method, pass through
Computer pattern recognition realizes the automatic identification of caste, while accuracy is high.
To achieve the above object, the technical solution that the present invention takes is:
Insect image identification detection method and classification of insect method, include the following steps:
S1, acquisition is converted to each frame image in video data to gray scale by colour, and by the gray level image with
Corresponding modeling background carries out difference operation;
S2, by the image segmentation of gained at a certain number of region units;
S3, the characteristic value for calculating each image block, the characteristic value includes brightness value, color feature value, direction character
Value, depth characteristic value and sparse eigenvalue;
S4, by each characteristic value of gained estimate linear normalization to [0,1] within the scope of;Each super-pixel unit is carried out
The fusion of four kinds of characteristic values obtains the Saliency maps of super-pixel precision;
S5, according to maximum variance between clusters, adaptive threshold fuzziness is carried out to the Saliency maps of the super-pixel precision of gained,
Obtain the binary map with prominent significant target part;
S6, SIFT feature is extracted from image different classes of in inquiry picture library using SIFT algorithms, by all features
Point vector set merges similar SIFT feature to one piece, using K-Means clustering algorithms, and construction one includes several words
The dictionary of remittance;
S7, extraction saliency region visual word, the number of visual word, constructs vision in statistical picture Saliency maps
Phrase generates the iamge description of image;
S8, each image in the iamge description of gained image and knowledge data base is subjected to images match, according to matched
As a result, judging the Known Species whether trypetid to be sorted belongs in knowledge data base.
Wherein, in the knowledge data base comprising known class a variety of trypetids characteristic.
Wherein, background modeling is carried out using averaging method.
Wherein, in the knowledge data base be equipped with a web crawlers, for search for the relevant website of inquired data or
Document.
Wherein, averaging method is that continuous N frames are taken in video image, and the average value for calculating this N frame image pixel gray level value is made
For the grey scale pixel value of background image.
Wherein, the threshold value value range in the step S5 is 0.165~0.315.
Wherein, knowledge data base is connected with a update module, for passing through 3G network, Wi-Fi network mode more new knowledge
Database.
The invention has the advantages that:
The automatic identification that caste is realized by computer pattern recognition, vision significance is fine with phrase
Combine, can more accurately reflect the salient region in query image, improve the accuracy of identification.
Specific implementation mode
In order to make objects and advantages of the present invention be more clearly understood, the present invention is carried out with reference to embodiments further
It is described in detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to limit this hair
It is bright.
An embodiment of the present invention provides a kind of insect image identification detection method and classification of insect methods, include the following steps:
S1, acquisition is converted to each frame image in video data to gray scale by colour, and by the gray level image with
Corresponding modeling background carries out difference operation;
S2, by the image segmentation of gained at a certain number of region units;
S3, the characteristic value for calculating each image block, the characteristic value includes brightness value, color feature value, direction character
Value, depth characteristic value and sparse eigenvalue;
S4, by each characteristic value of gained estimate linear normalization to [0,1] within the scope of;Each super-pixel unit is carried out
The fusion of four kinds of characteristic values obtains the Saliency maps of super-pixel precision;
S5, according to maximum variance between clusters, adaptive threshold fuzziness is carried out to the Saliency maps of the super-pixel precision of gained,
Obtain the binary map with prominent significant target part;
S6, SIFT feature is extracted from image different classes of in inquiry picture library using SIFT algorithms, by all features
Point vector set merges similar SIFT feature to one piece, using K-Means clustering algorithms, and construction one includes several words
The dictionary of remittance;
S7, extraction saliency region visual word, the number of visual word, constructs vision in statistical picture Saliency maps
Phrase generates the iamge description of image;
S8, each image in the iamge description of gained image and knowledge data base is subjected to images match, according to matched
As a result, judging the Known Species whether trypetid to be sorted belongs in knowledge data base.
The characteristic of a variety of trypetids comprising known class in the knowledge data base.
Background modeling is carried out using averaging method, is that pixel average is taken to some successive frames, basic thought is in video
Continuous N frames are taken in image, calculate grey scale pixel value of the average value of this N frame image pixel gray level value as background image.It is this
Algorithm speed is quickly.
It is equipped with a web crawlers in the knowledge data base, is used for search and the relevant website of inquired data or document,
Further improve the comprehensive of classification.
Threshold value value range in the step S5 is 0.165~0.315.
Knowledge data base is connected with a update module, for updating knowledge data by 3G network, Wi-Fi network mode
Library, can be with real-time update knowledge base data.
This specific implementation realizes the automatic identification of caste by computer pattern recognition, by vision significance
Combine well with phrase, can more accurately reflect the salient region in query image, improve the essence of identification
Exactness.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the principle of the present invention, it can also make several improvements and retouch, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (7)
1. insect image identification detection method and classification of insect method, which is characterized in that include the following steps:
S1, acquisition is converted to each frame image in video data to gray scale by colour, and by the gray level image with it is corresponding
Modeling background carry out difference operation;
S2, by the image segmentation of gained at a certain number of region units;
S3, the characteristic value for calculating each image block, the characteristic value include brightness value, color feature value, direction character value, depth
Spend characteristic value and sparse eigenvalue;
S4, by each characteristic value of gained estimate linear normalization to [0,1] within the scope of;Four kinds are carried out to each super-pixel unit
The fusion of characteristic value obtains the Saliency maps of super-pixel precision;
S5, the Saliency maps progress adaptive threshold fuzziness of the super-pixel precision of gained is obtained according to maximum variance between clusters
Binary map with the significant target part of protrusion;
S6, using SIFT algorithms from extracting SIFT feature in different classes of image in inquiry picture library, by all characteristic points to
Duration set merges similar SIFT feature to one piece, using K-Means clustering algorithms, and construction one includes several vocabulary
Dictionary;
S7, extraction saliency region visual word, the number of visual word, construction vision are short in statistical picture Saliency maps
Language generates the iamge description of image;
S8, each image in the iamge description of gained image and knowledge data base is subjected to images match, according to matched as a result,
Judge the Known Species whether trypetid to be sorted belongs in knowledge data base.
2. insect image identification detection method according to claim 1 and classification of insect method, which is characterized in that described knows
Know the characteristic of a variety of trypetids comprising known class in database.
3. insect image identification detection method according to claim 1 and classification of insect method, which is characterized in that utilize mean value
Method carries out background modeling.
4. insect image identification detection method according to claim 1 and classification of insect method, which is characterized in that the knowledge
A web crawlers is equipped in database, for searching for and the relevant website of inquired data or document.
5. insect image identification detection method according to claim 3 and classification of insect method, which is characterized in that averaging method is
Continuous N frames are taken in video image, calculate pixel grey scale of the average value of this N frame image pixel gray level value as background image
Value.
6. insect image identification detection method according to claim 1 and classification of insect method, which is characterized in that the step
Threshold value value range in S5 is 0.165~0.315.
7. insect image identification detection method according to claim 1 and classification of insect method, which is characterized in that knowledge data
Library is connected with a update module, for updating knowledge data base by 3G network, Wi-Fi network mode.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510638158.6A CN105243390B (en) | 2015-09-25 | 2015-09-25 | Insect image identification detection method and classification of insect method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510638158.6A CN105243390B (en) | 2015-09-25 | 2015-09-25 | Insect image identification detection method and classification of insect method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105243390A CN105243390A (en) | 2016-01-13 |
CN105243390B true CN105243390B (en) | 2018-09-25 |
Family
ID=55041031
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510638158.6A Expired - Fee Related CN105243390B (en) | 2015-09-25 | 2015-09-25 | Insect image identification detection method and classification of insect method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105243390B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107784020A (en) * | 2016-08-31 | 2018-03-09 | 司邦杰 | A kind of animals and plants insect species recognition methods |
CN106815819B (en) * | 2017-01-24 | 2019-08-13 | 河南工业大学 | More strategy grain worm visible detection methods |
CN107749066A (en) * | 2017-11-10 | 2018-03-02 | 深圳市唯特视科技有限公司 | A kind of multiple dimensioned space-time vision significance detection method based on region |
CN111046858B (en) * | 2020-03-18 | 2020-09-08 | 成都大熊猫繁育研究基地 | Image-based animal species fine classification method, system and medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102436656A (en) * | 2011-09-05 | 2012-05-02 | 同济大学 | Animal-diversity monitoring method based on computer vision |
CN102915446A (en) * | 2012-09-20 | 2013-02-06 | 复旦大学 | Plant disease and pest detection method based on SVM (support vector machine) learning |
CN104392240A (en) * | 2014-10-28 | 2015-03-04 | 中国疾病预防控制中心寄生虫病预防控制所 | Parasite egg identification method based on multi-feature fusion |
CN104680191A (en) * | 2014-11-24 | 2015-06-03 | 海南大学 | Active contour model-based detection method for plant pests and diseases |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8441489B2 (en) * | 2008-12-31 | 2013-05-14 | Intel Corporation | System and method for SIFT implementation and optimization |
-
2015
- 2015-09-25 CN CN201510638158.6A patent/CN105243390B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102436656A (en) * | 2011-09-05 | 2012-05-02 | 同济大学 | Animal-diversity monitoring method based on computer vision |
CN102915446A (en) * | 2012-09-20 | 2013-02-06 | 复旦大学 | Plant disease and pest detection method based on SVM (support vector machine) learning |
CN104392240A (en) * | 2014-10-28 | 2015-03-04 | 中国疾病预防控制中心寄生虫病预防控制所 | Parasite egg identification method based on multi-feature fusion |
CN104680191A (en) * | 2014-11-24 | 2015-06-03 | 海南大学 | Active contour model-based detection method for plant pests and diseases |
Non-Patent Citations (2)
Title |
---|
基于图像模式识别技术的实蝇昆虫分类识别研究;郝中华;《中国优秀硕士学位论文全文数据库 信息科技辑》;20120229;I138-1764 * |
实蝇昆虫图像特征提取与识别技术的研究;李正;《中国优秀硕士学位论文全文数据库 农业科技辑》;20120531;D046-2 * |
Also Published As
Publication number | Publication date |
---|---|
CN105243390A (en) | 2016-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111178197B (en) | Mass R-CNN and Soft-NMS fusion based group-fed adherent pig example segmentation method | |
CN111738064B (en) | Haze concentration identification method for haze image | |
CN106295653B (en) | Water quality image classification method | |
CN112308158A (en) | Multi-source field self-adaptive model and method based on partial feature alignment | |
CN112241762B (en) | Fine-grained identification method for pest and disease damage image classification | |
CN105243390B (en) | Insect image identification detection method and classification of insect method | |
CN111540006B (en) | Plant stomata intelligent detection and identification method and system based on deep migration learning | |
CN104778711B (en) | A kind of cell count localization method of cleavage stage | |
CN110097060B (en) | Open set identification method for trunk image | |
CN104778481A (en) | Method and device for creating sample library for large-scale face mode analysis | |
CN109034269A (en) | A kind of bollworm female male imago method of discrimination based on computer vision technique | |
CN109063619A (en) | A kind of traffic lights detection method and system based on adaptive background suppression filter and combinations of directions histogram of gradients | |
Travieso et al. | Pollen classification based on contour features | |
CN109214298A (en) | A kind of Asia women face value Rating Model method based on depth convolutional network | |
CN103177266A (en) | Intelligent stock pest identification system | |
CN114463843A (en) | Multi-feature fusion fish abnormal behavior detection method based on deep learning | |
CN114299279B (en) | Mark-free group rhesus monkey motion quantity estimation method based on face detection and recognition | |
CN106326914B (en) | A kind of more classification methods of pearl based on SVM | |
Utaminingrum et al. | Alphabet Sign Language Recognition Using K-Nearest Neighbor Optimization. | |
CN112257730A (en) | Plant pest image identification method, device, equipment and storage medium | |
CN111191510A (en) | Relation network-based remote sensing image small sample target identification method in complex scene | |
CN113283378B (en) | Pig face detection method based on trapezoidal region normalized pixel difference characteristics | |
CN113553947B (en) | Method and device for generating and describing multi-mode pedestrian re-recognition and electronic equipment | |
CN111046861A (en) | Method for identifying whether sample equipment exists in infrared image, method for constructing power equipment identification model and application | |
Li et al. | Classification of Diospyros Lotus seeds based on MLP neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB03 | Change of inventor or designer information |
Inventor after: Qin Xuefeng Inventor after: Mao Xiaorui Inventor after: Zhang Lilei Inventor after: Yang Helian Inventor before: Qin Xuefeng Inventor before: Li Weihai Inventor before: Zhang Lilei Inventor before: Yang Helian |
|
CB03 | Change of inventor or designer information | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180925 Termination date: 20190925 |
|
CF01 | Termination of patent right due to non-payment of annual fee |