CN103148781A - Grapefruit size estimating method based on binocular vision - Google Patents
Grapefruit size estimating method based on binocular vision Download PDFInfo
- Publication number
- CN103148781A CN103148781A CN2013100439584A CN201310043958A CN103148781A CN 103148781 A CN103148781 A CN 103148781A CN 2013100439584 A CN2013100439584 A CN 2013100439584A CN 201310043958 A CN201310043958 A CN 201310043958A CN 103148781 A CN103148781 A CN 103148781A
- Authority
- CN
- China
- Prior art keywords
- shaddock
- image
- images
- width
- grapefruit
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a grapefruit size estimating method based on binocular vision. Grapefruit images are obtained by using two shooting angles, wherein included angles of the two shooting angles are ninety degrees. The grapefruit images are segmented and pixel points in the images are obtained. Corresponding cross section areas of the grapefruit images are obtained. Grapefruit image capture conditions are improved and accurate estimation to grapefruit sizes are achieved by utilizing a neural network. Detecting efficiency and accuracy are high. Former problems are well solved. The grapefruit size estimating method based on the binocular vision is strong in practicability, convenient to operate and capable of having good values of popularization and application.
Description
Technical field
The invention belongs to volume of fruits estimation technical field, relate in particular to a kind of shaddock based on binocular vision, Navel Orange Fruits volume estimating and measuring method.
Background technology
In existing fruit automatic classification, the volume of fruits size is an important technical indicator, and main dependence is manually carried out in existing shaddock classification, inefficiency, poor accuracy.For this phenomenon, introduce machine vision and carry out the developing direction that automatic classification becomes the shaddock classification, but because obtaining picture quality can't accurately reflect volume of fruits, instead index is carried out shaddock to general selection fruit sectional area, Navel Orange Fruits carries out classification, but in the method that existing fruit vision detects, lack the accurate estimating and measuring method for volume of fruits, mostly the volume of fruits size is divided into some grades and carries out classification.
Summary of the invention
The invention provides a kind of shaddock volume estimating and measuring method based on binocular vision, being intended to solution main dependence in existing shaddock classification manually carries out, inefficiency, poor accuracy, and because picture quality that machine vision is obtained can't accurately reflect volume of fruits, instead index is carried out shaddock to general selection fruit sectional area, Navel Orange Fruits carries out classification, poor accuracy can't accurately reflect the problem of the volume of fruit.
The object of the present invention is to provide a kind of shaddock volume estimating and measuring method based on binocular vision, this shaddock volume estimating and measuring method comprises the following steps:
Step 1 is obtained the shaddock image,
Step 2, and respectively two width images are carried out image segmentation, obtain the pixel of shaddock in image;
Step 3 with the height of shaddock in the difference presentation video of shaddock pixel corresponding row coordinate maximal value and minimum value in image, is calculated respectively the height of shaddock in two width images;
Step 4 is adjusted the size of the less shaddock image of height value and corresponding reference substance image by interpolation, make after adjusting that in two width images, the height of shaddock equates;
Step 5, according to the every row pixel of object of reference sectional area number in object of reference image after adjusting, two sectional areas corresponding to width shaddock image every row pixel number after calculating is adjusted respectively;
Step 6, after adjusting, the corresponding sectional area summation of all row of two width shaddock images, obtain two sectional areas corresponding to width shaddock image;
Step 7 take sectional area corresponding to the two width shaddock images that obtain as input, adopts neural network to carry out the volume estimation of shaddock.
Further, obtaining of described shaddock image is to be the shooting of 90 ° from two angles.
shaddock volume estimating and measuring method based on binocular vision provided by the invention, utilizing two angles is that the shooting angle of 90 ° is obtained the shaddock image, and to the shaddock image segmentation, obtain the pixel in image, obtain sectional area corresponding to shaddock image, improve shaddock image acquisition condition, and adopt neural network to realize the accurate estimation of shaddock volume, detection efficiency and accuracy are high, solved because picture quality that machine vision is obtained can't accurately reflect volume of fruits, general selection fruit sectional area instead index carries out the problem that shaddock carries out classification, practical, having stronger propagation and employment is worth.
Description of drawings
Fig. 1 is the realization flow figure based on the shaddock volume estimating and measuring method of binocular vision that the embodiment of the present invention provides;
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is described in further detail.Should be appreciated that specific embodiment described herein only in order to explaining the present invention, and be not used in and limit invention.
The realization flow based on the shaddock volume estimating and measuring method of binocular vision that Fig. 1 shows that the embodiment of the present invention provides.
This shaddock volume estimating and measuring method comprises the following steps:
Step S101 obtains the shaddock image;
Step S102, and respectively two width images are carried out image segmentation, obtain the pixel of shaddock in image;
Step S103 with the height of shaddock in the difference presentation video of shaddock pixel corresponding row coordinate maximal value and minimum value in image, calculates respectively the height of shaddock in two width images;
Step S104 adjusts the size of the less shaddock image of height value and corresponding reference substance image by interpolation, make after adjusting that in two width images, the height of shaddock equates;
Step S105, according to the every row pixel of object of reference sectional area number in object of reference image after adjusting, two sectional areas corresponding to width shaddock image every row pixel number after calculating is adjusted respectively;
Step S106, after adjusting, the corresponding sectional area summation of all row of two width shaddock images, obtain two sectional areas corresponding to width shaddock image;
Step S107 take sectional area corresponding to the two width shaddock images that obtain as input, adopts neural network to carry out the volume estimation of shaddock.
In embodiments of the present invention, obtaining of shaddock image is to be the shooting of 90 ° from two angles.
In embodiments of the present invention, this shaddock volume estimating and measuring method can be used for the volume of shaddock is estimated.
Below in conjunction with drawings and the specific embodiments, application principle of the present invention is further described.
Describe as an example of the estimation of shaddock volume example, as shown in Figure 1, this method estimation shaddock volume is divided into 6 steps:; 1) be that the shooting angle of 90 ° is obtained the shaddock image from two angles; 2) also respectively two width images are carried out image segmentation, obtain the pixel of shaddock in image; 3) with the height of shaddock in the difference presentation video of shaddock pixel corresponding row coordinate maximal value and minimum value in image, calculate respectively the height of shaddock in two width images; 4) adjust the size of the less shaddock image of height value and corresponding reference substance image by interpolation, make the height of adjusting shaddock in rear two width images equate; 5) according to the every row pixel of object of reference sectional area number in object of reference image after adjusting, two sectional areas corresponding to width shaddock image every row pixel number after calculating is adjusted respectively; 6) will adjust the corresponding sectional areas summation of rear all row of two width shaddock images, obtain two sectional areas corresponding to width shaddock image; 7) take sectional area corresponding to the two width shaddock images that obtain as input, adopt neural network to carry out the volume estimation of shaddock; Wherein, the experimental result error of volume estimation is less than 5%.
the shaddock volume estimating and measuring method based on binocular vision that the embodiment of the present invention provides, utilizing two angles is that the shooting angle of 90 ° is obtained the shaddock image, and to the shaddock image segmentation, obtain the pixel in image, obtain sectional area corresponding to shaddock image, improve shaddock image acquisition condition, and adopt neural network to realize the accurate estimation of shaddock volume, detection efficiency and accuracy are high, solved because picture quality that machine vision is obtained can't accurately reflect volume of fruits, general selection fruit sectional area instead index carries out the problem that shaddock carries out classification, practical, having stronger propagation and employment is worth.
The above is only preferred embodiment of the present invention, not in order to limiting the present invention, all any modifications of doing within the spirit and principles in the present invention, is equal to and replaces and improvement etc., within all should being included in protection scope of the present invention.
Claims (2)
1. the shaddock volume estimating and measuring method based on binocular vision, is characterized in that, described shaddock volume estimating and measuring method comprises the following steps:
Step 1 is obtained the shaddock image,
Step 2, and respectively two width images are carried out image segmentation, obtain the pixel of shaddock in image;
Step 3 with the height of shaddock in the difference presentation video of shaddock pixel corresponding row coordinate maximal value and minimum value in image, is calculated respectively the height of shaddock in two width images;
Step 4 is adjusted the size of the less shaddock image of height value and corresponding reference substance image by interpolation, make after adjusting that in two width images, the height of shaddock equates;
Step 5, according to the every row pixel of object of reference sectional area number in object of reference image after adjusting, two sectional areas corresponding to width shaddock image every row pixel number after calculating is adjusted respectively;
Step 6, after adjusting, the corresponding sectional area summation of all row of two width shaddock images, obtain two sectional areas corresponding to width shaddock image;
Step 7 take sectional area corresponding to the two width shaddock images that obtain as input, adopts neural network to carry out the volume estimation of shaddock.
2. the shaddock volume estimating and measuring method based on binocular vision as claimed in claim 1, is characterized in that, obtaining of described shaddock image is to be the shooting of 90 ° from two angles.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100439584A CN103148781A (en) | 2013-01-26 | 2013-01-26 | Grapefruit size estimating method based on binocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100439584A CN103148781A (en) | 2013-01-26 | 2013-01-26 | Grapefruit size estimating method based on binocular vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103148781A true CN103148781A (en) | 2013-06-12 |
Family
ID=48546994
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100439584A Pending CN103148781A (en) | 2013-01-26 | 2013-01-26 | Grapefruit size estimating method based on binocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103148781A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103486967A (en) * | 2013-09-02 | 2014-01-01 | 中国农业科学院农产品加工研究所 | Method for rapidly measuring fruit volume |
| CN104764402A (en) * | 2015-03-11 | 2015-07-08 | 广西科技大学 | Visual inspection method for citrus size |
| CN106839977A (en) * | 2016-12-23 | 2017-06-13 | 西安科技大学 | Shield dregs volume method for real-time measurement based on optical grating projection binocular imaging technology |
| CN108038879A (en) * | 2017-12-12 | 2018-05-15 | 众安信息技术服务有限公司 | A kind of volume of food method of estimation and its device |
| CN109509535A (en) * | 2018-10-08 | 2019-03-22 | 北京健康有益科技有限公司 | The acquisition methods of food volume, the acquisition methods of fuel value of food, electronic equipment |
| CN110060294A (en) * | 2019-04-30 | 2019-07-26 | 中国农业科学院农业环境与可持续发展研究所 | A kind of yield assessment method of fruit tree crop |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0178090A1 (en) * | 1984-09-19 | 1986-04-16 | Ishida Scales Mfg. Co. Ltd. | Volume determination process |
| JP2007064939A (en) * | 2005-09-02 | 2007-03-15 | Mitsui Mining & Smelting Co Ltd | Peel puffing discrimination apparatus of vegetables and fruits and peel puffing discrimination method of vegetables and fruits |
| CN102680414A (en) * | 2012-05-30 | 2012-09-19 | 华中农业大学 | Automatic grading device for red globe grapes based on machine vision and method thereof |
-
2013
- 2013-01-26 CN CN2013100439584A patent/CN103148781A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0178090A1 (en) * | 1984-09-19 | 1986-04-16 | Ishida Scales Mfg. Co. Ltd. | Volume determination process |
| JP2007064939A (en) * | 2005-09-02 | 2007-03-15 | Mitsui Mining & Smelting Co Ltd | Peel puffing discrimination apparatus of vegetables and fruits and peel puffing discrimination method of vegetables and fruits |
| CN102680414A (en) * | 2012-05-30 | 2012-09-19 | 华中农业大学 | Automatic grading device for red globe grapes based on machine vision and method thereof |
Non-Patent Citations (3)
| Title |
|---|
| 胡波等: "基于机器视觉脐橙体积的球体估测方法", 《安徽农业科学》 * |
| 赵攀等: "基于图像参数的BP网络之岩石颗粒体积估算", 《计算机测量与控制》 * |
| 黄玲等: "基于机器视觉的脐橙可食率的研究", 《安徽农业科学》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103486967A (en) * | 2013-09-02 | 2014-01-01 | 中国农业科学院农产品加工研究所 | Method for rapidly measuring fruit volume |
| CN104764402A (en) * | 2015-03-11 | 2015-07-08 | 广西科技大学 | Visual inspection method for citrus size |
| CN106839977A (en) * | 2016-12-23 | 2017-06-13 | 西安科技大学 | Shield dregs volume method for real-time measurement based on optical grating projection binocular imaging technology |
| CN106839977B (en) * | 2016-12-23 | 2019-05-07 | 西安科技大学 | Shield dregs volume method for real-time measurement based on optical grating projection binocular imaging technology |
| CN108038879A (en) * | 2017-12-12 | 2018-05-15 | 众安信息技术服务有限公司 | A kind of volume of food method of estimation and its device |
| CN109509535A (en) * | 2018-10-08 | 2019-03-22 | 北京健康有益科技有限公司 | The acquisition methods of food volume, the acquisition methods of fuel value of food, electronic equipment |
| CN110060294A (en) * | 2019-04-30 | 2019-07-26 | 中国农业科学院农业环境与可持续发展研究所 | A kind of yield assessment method of fruit tree crop |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103148781A (en) | Grapefruit size estimating method based on binocular vision | |
| CN108615244B (en) | A kind of image depth estimation method and system based on CNN and depth filter | |
| CN105528785B (en) | A kind of binocular vision image solid matching method | |
| CN107844757B (en) | Method for extracting river channel width in river basin by using digital elevation model | |
| CN102436652B (en) | Automatic registering method of multisource remote sensing images | |
| Wieneke | Improvements for volume self-calibration | |
| CN104135597B (en) | A kind of video jitter automatic testing method | |
| CN102982537B (en) | A kind of method and system detecting scene change | |
| CN106952274B (en) | Pedestrian detection and distance measuring method based on stereoscopic vision | |
| CN105793892B (en) | A kind of image processing method, device and picture pick-up device | |
| CN103606132A (en) | Multiframe digital image denoising method based on space domain and time domain combination filtering | |
| CN104392435A (en) | Fisheye camera calibration method and device | |
| CN108765480B (en) | Advanced treatment equipment | |
| CN108088381A (en) | A kind of contactless minim gap method for measuring width based on image procossing | |
| CN104794717A (en) | Binocular vision system based depth information comparison method | |
| US10157439B2 (en) | Systems and methods for selecting an image transform | |
| CN104881866A (en) | Fisheye camera rectification and calibration method for expanding pin-hole imaging model | |
| CN105139355A (en) | Method for enhancing depth images | |
| CN110345875A (en) | Calibration and distance measuring method, device, electronic equipment and computer readable storage medium | |
| KR101868483B1 (en) | Prediction of two blur parameters of edges in varying contrast | |
| CN104200453A (en) | Parallax image correcting method based on image segmentation and credibility | |
| WO2019048904A1 (en) | Combined stereoscopic and phase detection depth mapping in a dual aperture camera | |
| CN108805940A (en) | A kind of fast algorithm of zoom camera track and localization during zoom | |
| CN106482648B (en) | The absolute monitoring device of thin tail sheep and method in long-distance plane based on fixed point | |
| CN103983239B (en) | Based on the distance-finding method of the wide line in track |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| ASS | Succession or assignment of patent right |
Owner name: GUANGXI UNIVERSITY OF TECHNOLOGY Free format text: FORMER OWNER: LUSHAN COLLEGE GUANGXI UNIVERSITY OF SCIENCE AND TECHNOLOGY Effective date: 20130708 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| C53 | Correction of patent of invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Hu Bo Inventor after: Ma Zhaomin Inventor after: Huang Ling Inventor after: Shi Yuqiu Inventor after: Liu Ruiqi Inventor after: Qin Yongxin Inventor before: Cao Naiwen Inventor before: Hu Bo |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: CAO NAIWEN HU BO TO: HU BO MA ZHAOMIN HUANG LING SHI YUQIU LIU RUIQI QIN YONGXIN Free format text: CORRECT: ADDRESS; FROM: 545616 LIUZHOU, GUANGXI ZHUANG AUTONOMOUS REGION TO: 545000 LIUZHOU, GUANGXI ZHUANG AUTONOMOUS REGION |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20130708 Address after: 545000 No. 268 East Ring Road, the Guangxi Zhuang Autonomous Region, Liuzhou Applicant after: Guangxi University of Technology Address before: 545616 Liu Dong Road, willow New District, the Guangxi Zhuang Autonomous Region, Liuzhou Applicant before: Lushan College, Guangxi University of Science and Technology |
|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130612 |