CN103714317A - Plant blade recognition method based on Android platform - Google Patents
Plant blade recognition method based on Android platform Download PDFInfo
- Publication number
- CN103714317A CN103714317A CN201310670473.8A CN201310670473A CN103714317A CN 103714317 A CN103714317 A CN 103714317A CN 201310670473 A CN201310670473 A CN 201310670473A CN 103714317 A CN103714317 A CN 103714317A
- Authority
- CN
- China
- Prior art keywords
- plant
- image
- android platform
- detected
- identification
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a plant blade recognition method based on an Android platform. The method comprises: (1) performing tailoring of an interested area on an image of a plant to be detected, executing (2) if selecting stand-alone recognition, and executing (4) if selecting networking recognition; (2) in cooperation with a K nearest node classifier and a support vector machine, a database of the Android platform is used to perform classification recognition on the image of the plant to be detected; (3) getting the Euclidean distance of a feature description vector of the image of the plant to be detected and feature description vectors of all plant images in the database, taking the first five plant species with the smallest Euclidean distance, judging whether the obtained target plant species is among the five plant species, executing (5) if yes, and executing (4) otherwise; and (4) sending an interested image to a server side through an HTTP protocol. Compared with the prior art, the plant blade recognition method based on the Android platform processes images in real time and efficiently, and uses NDK and JNI technologies to remarkably improve the accuracy and the speed of recognition of classification.
Description
Technical field
The present invention relates to Digital Image Processing and mode identification technology, especially relate to a kind of plant leaf identification method based on Android platform.
Background technology
Nearly 400000 Plants on the earth, China is second-biggest-in-the-world plant species resources bank.The classification of plant species is the important foundation work of botany research and agricultural production with identification, it is human knowledge plant, grasp the evolution laws of plant, understand the fundamental characteristics of plant and the basis being used, can be applicable to the fields such as diseases and pests of agronomic crop identification, crop growth monitoring, simultaneously with agriculture, woods, herd, also close relation of the field such as pair, fishing, traditional Chinese medicine.Traditional Plant Taxonomy study general needs manual measurement, obtains data the sample gathering, the surface of comprehensive plant, classifies by people's common sense, and this method workload is huge, and subjective, cause the possibility of identification error large.Along with the fast development of the technology such as machine learning, pattern-recognition, image processing, the powerful calculating ability that we can computer, carries out quick and objective classification to plant.In the root of plant, stem, leaf, flower, fruit, the large organ of seed six, leaf is relatively large with its quantity, and growth cycle is relatively short, and gathering the advantage such as convenient, injury is less becomes the first-selection of carrying out plant identification.Due to the difference of growing environment, the difference of gene, the blade of different plants has either large or small difference on shape, texture, therefore gathers the image of blade, obtains its characteristic parameter, will make people classify accurately to plant.
Modern plant leaf blade recognition system is by obtaining the leaf image that quality is higher, image is carried out to morphologic pre-service, be partitioned into interested part, then by efficient algorithm, extract the characteristic parameter of image, finally characteristic parameter is sent into sorter, obtain result.Be about to image processing and be applied on leaf image with the theory and technology of machine learning, reach the object of classification and identification.
Past 10 years, the countries such as America and Europe constantly appeal to adopt Digital Image Processing and mode identification technology that plant species is identified and realized full-automatic or semi-automatic computer aided calculation, the people such as Milanie Thibaut are in the sort research that Salicaceous Plants is carried out, and the people such as T.Saitoh identify the plant of wild species at work with the image of flower and blade.Only have in the world at present Columbia Univ USA and Smith's botany can be in exploitation the mobile phone based on IOS system and the plant leaf blade recognition system of panel computer.And the domestic plant leaf blade identification software there is no based on mobile device.
Rise along with Android system, a large amount of mobile devices comprises that mobile phone, panel computer have all adopted this powerful embedded OS, in November, 2012, data showed, Android occupies the share in global operation system of smart phone market 76%, and Chinese market occupation rate is 90%.Because the growing environment of plant is substantially all outdoor, traditional desktop computer, notebook is not suitable for carrying to gather identification plant leaf blade, and handheld device volume is little, be convenient for carrying, and the arrival in the universal and 3G epoch of camera system, will extremely be conducive to utilize mobile phone, flat board etc. to gather blade, by the form of unit or networking, plant leaf blade be carried out to processing and identification.
Summary of the invention
Object of the present invention is exactly to provide a kind of plant leaf identification method based on Android platform in order to overcome the defect of above-mentioned prior art existence.
Object of the present invention can be achieved through the following technical solutions: a kind of plant leaf identification method based on Android platform, it is characterized in that, and comprise the following steps:
(1) utilize the image of Android platform to process API, plant image to be detected is carried out to the cutting of Gan Xing district, cutting Hou Ganxing district image is selected to execution step (2) of unit identification, select execution step (4) of networking identification;
(2) calculate the character gradient histogram feature vector of Gan Xing district image, the neighbouring node sorter of associating K and support vector machine, utilize the database of Android platform to carry out Classification and Identification to plant image to be detected, obtain the affiliated target plant kind of plant to be detected;
(3) ask for the Euclidean distance of the feature description vectors of all plant images in the feature description vectors of plant image to be detected and the database of Android platform, get the first five Plants kind of Euclidean distance minimum, whether the target plant kind that determining step (2) obtains is in five Plants kinds of Euclidean distance minimum, to perform step (5), otherwise, execution step (4);
(4) by http protocol Jiang Ganxing district image, send to server end, server end receives with servlet, afterwards the image receiving is extracted to character gradient histogram feature vector, then send into support vector machine and carry out Classification and Identification, and recognition result is returned to Android platform;
(5) show root, stem, leaf, flower, fruit and the seed picture of plant to be detected, and the science popularization recommended information content of this Plants.
Implementation step on Android platform (2) is specially, build Android development environment, download and install configuration NDK and Cygwin, utilize afterwards JNI to write the histograms of oriented gradients new image representation method file of the neighbouring node sorter of associating K and support vector machine, recycling NDK generates dynamic link library, the method file packing of writing is placed in Android platform, and operation method file is treated detected image and is carried out Classification and Identification.
Compared with prior art, the inventive method can be applicable to the terminal based on Android platform, realization is carried out analyzing and processing to the leaf image collecting, by image acquisition, cut apart, the function such as displaying, classification, storage, retrieval is integrated in one, method clear in structure, it is real-time, efficient that image is processed, and utilizes NDK and JNI technology simultaneously, significantly improved accuracy and the speed of recognition and classification.
Embodiment
Below in conjunction with specific embodiment, the present invention is described in detail.
A plant leaf identification method based on Android platform, is characterized in that, comprises the following steps:
(1) utilize the image of Android platform to process API, plant image to be detected is carried out to the cutting of Gan Xing district, cutting Hou Ganxing district image is selected to execution step (2) of unit identification, select execution step (4) of networking identification;
(2) calculate the character gradient histogram feature vector of Gan Xing district image, the neighbouring node sorter of associating K and support vector machine, utilize the database of Android platform to carry out Classification and Identification to plant image to be detected, obtain the affiliated target plant kind of plant to be detected;
(3) ask for the Euclidean distance of the feature description vectors of all plant images in the feature description vectors of plant image to be detected and the database of Android platform, get the first five Plants kind of Euclidean distance minimum, whether the target plant kind that determining step (2) obtains is in five Plants kinds of Euclidean distance minimum, to perform step (5), otherwise, execution step (4);
(4) by http protocol Jiang Ganxing district image, send to server end, server end receives with servlet, afterwards the image receiving is extracted to character gradient histogram feature vector, then send into support vector machine and carry out Classification and Identification, and recognition result is returned to Android platform;
(5) show root, stem, leaf, flower, fruit and the seed picture of plant to be detected, and the science popularization recommended information content of this Plants.
Implementation step on Android platform (2) is specially, build Android development environment, download and install configuration NDK and Cygwin, utilize afterwards JNI to write the histograms of oriented gradients new image representation method file of the neighbouring node sorter of associating K and support vector machine, recycling NDK generates dynamic link library, the method file packing of writing is placed in Android platform, and operation method file is treated detected image and is carried out Classification and Identification.
Claims (2)
1. the plant leaf identification method based on Android platform, is characterized in that, comprises the following steps:
(1) utilize the image of Android platform to process API, plant image to be detected is carried out to the cutting of Gan Xing district, cutting Hou Ganxing district image is selected to execution step (2) of unit identification, select execution step (4) of networking identification;
(2) calculate the character gradient histogram feature vector of Gan Xing district image, the neighbouring node sorter of associating K and support vector machine, utilize the database of Android platform to carry out Classification and Identification to plant image to be detected, obtain the affiliated target plant kind of plant to be detected;
(3) ask for the Euclidean distance of the feature description vectors of all plant images in the feature description vectors of plant image to be detected and the database of Android platform, get the first five Plants kind of Euclidean distance minimum, whether the target plant kind that determining step (2) obtains is in five Plants kinds of Euclidean distance minimum, to perform step (5), otherwise execution step (4);
(4) by http protocol Jiang Ganxing district image, send to server end, server end receives with servlet, afterwards the image receiving is extracted to character gradient histogram feature vector, then send into support vector machine and carry out Classification and Identification, and recognition result is returned to Android platform;
(5) show root, stem, leaf, flower, fruit and the seed picture of plant to be detected, and the science popularization recommended information content of this Plants.
2. a kind of plant leaf identification method based on Android platform according to claim 1, it is characterized in that, implementation step on Android platform (2) is specially, build Android development environment, download and install configuration NDK and Cygwin, utilize afterwards JNI to write the histograms of oriented gradients new image representation method file of the neighbouring node sorter of associating K and support vector machine, recycling NDK generates dynamic link library, the method file packing of writing is placed in Android platform, and operation method file is treated detected image and is carried out Classification and Identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310670473.8A CN103714317A (en) | 2013-12-10 | 2013-12-10 | Plant blade recognition method based on Android platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310670473.8A CN103714317A (en) | 2013-12-10 | 2013-12-10 | Plant blade recognition method based on Android platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103714317A true CN103714317A (en) | 2014-04-09 |
Family
ID=50407277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310670473.8A Pending CN103714317A (en) | 2013-12-10 | 2013-12-10 | Plant blade recognition method based on Android platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103714317A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036235A (en) * | 2014-05-27 | 2014-09-10 | 同济大学 | Plant species identification method based on leaf HOG features and intelligent terminal platform |
| CN104331713A (en) * | 2014-11-04 | 2015-02-04 | 无锡北斗星通信息科技有限公司 | Mobile terminal for detecting leaf varieties |
| CN104915682A (en) * | 2015-06-08 | 2015-09-16 | 浪潮集团有限公司 | Leguminous seed identification system and method |
| CN105203456A (en) * | 2015-10-28 | 2015-12-30 | 小米科技有限责任公司 | Plant species identification method and apparatus thereof |
| CN105631451A (en) * | 2016-01-07 | 2016-06-01 | 同济大学 | Plant leave identification method based on android system |
| CN108073947A (en) * | 2017-04-10 | 2018-05-25 | 大连大学 | A kind of method for identifying blueberry kind |
| CN108898059A (en) * | 2018-05-30 | 2018-11-27 | 上海应用技术大学 | Flowers recognition methods and its equipment |
-
2013
- 2013-12-10 CN CN201310670473.8A patent/CN103714317A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036235A (en) * | 2014-05-27 | 2014-09-10 | 同济大学 | Plant species identification method based on leaf HOG features and intelligent terminal platform |
| CN104036235B (en) * | 2014-05-27 | 2017-07-07 | 同济大学 | Plant species recognition methods based on blade HOG features and intelligent terminal platform |
| CN104331713A (en) * | 2014-11-04 | 2015-02-04 | 无锡北斗星通信息科技有限公司 | Mobile terminal for detecting leaf varieties |
| CN104331713B (en) * | 2014-11-04 | 2017-07-07 | 许金普 | Mobile terminal for detecting leaf kind |
| CN104915682A (en) * | 2015-06-08 | 2015-09-16 | 浪潮集团有限公司 | Leguminous seed identification system and method |
| CN105203456A (en) * | 2015-10-28 | 2015-12-30 | 小米科技有限责任公司 | Plant species identification method and apparatus thereof |
| CN105631451A (en) * | 2016-01-07 | 2016-06-01 | 同济大学 | Plant leave identification method based on android system |
| CN108073947A (en) * | 2017-04-10 | 2018-05-25 | 大连大学 | A kind of method for identifying blueberry kind |
| CN108073947B (en) * | 2017-04-10 | 2021-08-27 | 大连大学 | Method for identifying blueberry varieties |
| CN108898059A (en) * | 2018-05-30 | 2018-11-27 | 上海应用技术大学 | Flowers recognition methods and its equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103714317A (en) | Plant blade recognition method based on Android platform | |
| Liu et al. | Tomato diseases and pests detection based on improved Yolo V3 convolutional neural network | |
| Tian et al. | Detection of apple lesions in orchards based on deep learning methods of CycleGAN and YOLOV3‐dense | |
| Song et al. | Kiwifruit detection in field images using Faster R-CNN with VGG16 | |
| Lin et al. | Quick and accurate monitoring peanut seedlings emergence rate through UAV video and deep learning | |
| Sun et al. | Image processing algorithms for infield single cotton boll counting and yield prediction | |
| CN113657469B (en) | Automatic observation method and system for woody plant waiting period based on image recognition | |
| Liu et al. | The recognition of apple fruits in plastic bags based on block classification | |
| CN107680098A (en) | A kind of recognition methods of sugarcane sugarcane section feature | |
| CN110751035A (en) | Seed corn production identification method and device | |
| Wang et al. | A review on the application of computer vision and machine learning in the tea industry | |
| Kumar et al. | Analysis of apple plant leaf diseases detection and classification: A review | |
| Wang et al. | TBC-YOLOv7: a refined YOLOv7-based algorithm for tea bud grading detection | |
| Li et al. | Disease recognition of maize leaf based on KNN and feature extraction | |
| CN110060294A (en) | A kind of yield assessment method of fruit tree crop | |
| Zhao et al. | Transient multi-indicator detection for seedling sorting in high-speed transplanting based on a lightweight model | |
| Huang et al. | A survey of deep learning-based object detection methods in crop counting | |
| Zhang et al. | Automatic non-destructive multiple lettuce traits prediction based on DeepLabV3+ | |
| CN105631451A (en) | Plant leave identification method based on android system | |
| CN203414962U (en) | Intelligent terminal-based plant leaf recognition device | |
| CN103824083A (en) | Web online species recognition method based on blade complete and partial two-value characteristics | |
| Jin et al. | Hydroponic lettuce defective leaves identification based on improved YOLOv5s | |
| Zhang et al. | A tea bud segmentation, detection and picking point localization based on the MDY7-3PTB model | |
| Yihang et al. | Automatic recognition of rape seeding emergence stage based on computer vision technology | |
| Guo et al. | Automatic and accurate acquisition of stem-related phenotypes of mature soybean based on deep learning and directed search algorithms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140409 |
|
| RJ01 | Rejection of invention patent application after publication |