Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Contour-based target fruit image instance segmentation method and system

A target and fruit technology, applied in the field of image processing, can solve the problems of insufficient fruit picking speed, resource and memory occupation, and insufficient speed, etc., and achieve the effect of fast segmentation speed, small amount of calculation, and fast speed

Pending Publication Date: 2021-08-06
SHANDONG NORMAL UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the inventors found that these fruit recognition methods are often segmented based on pixels. The pixel-based segmentation method has a large amount of calculation and is not fast enough, resulting in taking up a lot of resources and memory, and the speed of fruit picking cannot reach reality. purpose requirements

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Contour-based target fruit image instance segmentation method and system
  • Contour-based target fruit image instance segmentation method and system
  • Contour-based target fruit image instance segmentation method and system

Examples

Experimental program
Comparison scheme
Effect test

example

[0050] Such as figure 1 As shown, the present embodiment provides a contour-based target fruit image instance segmentation method, which specifically includes the following steps:

[0051] Step S101: Obtain an image of the target fruit.

[0052] In specific implementation, the target fruit image acquired can be the image of any angle, time period and weather. Moreover, when actually training the first target detection network and the second target detection network, the equipment is also used to collect images of target fruits (such as apples) and shoot them at different angles, time periods and weather to enrich the data set, such as Figure 5(a)-Figure 5(f) shown. And use the labelme software to label, and make the labeled fruits into a data set.

[0053] Step S102: extracting features of the target fruit image, and performing feature fusion.

[0054] Before extracting the features of the target fruit image, pre-process the collected image such as smoothing and noise red...

Embodiment 2

[0079] The present embodiment provides a kind of contour-based target fruit image instance segmentation system, which specifically includes the following modules:

[0080] Image acquisition module, which is used to acquire target fruit image;

[0081] Feature fusion module, it is used to extract the feature of target fruit image, and carries out feature fusion;

[0082] The first diagonal key point detection module is used to input the fusion feature to the first target detection network to obtain the first diagonal key point coordinates and a candidate frame;

[0083] The second diagonal key point detection module, which is used to input the image in the candidate frame to the second target detection network, obtains at least two second diagonal key point coordinates and corresponding candidate frames;

[0084] A contour location module, which is used to construct an initial contour based on the second diagonal key point coordinates and corresponding candidate frames, deform...

Embodiment 3

[0088] The present embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in the contour-based target fruit image instance segmentation method as described above are realized.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the field of image processing, and provides a contour-based target fruit image instance segmentation method and system. The method comprises the following steps: acquiring a target fruit image; extracting features of the target fruit image, and performing feature fusion; inputting the fusion feature into a first target detection network to obtain a first diagonal key point coordinate and a candidate box; inputting the images in the candidate frames into a second target detection network to obtain at least two second diagonal key point coordinates and corresponding candidate frames; and constructing an initial contour based on the second diagonal key point coordinates and the corresponding candidate frame, deforming the initial contour to a target boundary, determining a final contour of the target fruit, and segmenting positions of all target fruits in the target fruit image.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a method and system for segmenting target fruit image instances based on contours. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] With the rise of modern artificial intelligence, the level of modern agricultural technology has been greatly improved, especially the rise of image recognition technology, which has brought huge benefits to agricultural production. Traditional manual labor has gradually been canceled and replaced by robotic operations. Such as robot picking, robot drug spraying, robot output estimation, etc. The advancement of image recognition technology provides strong technical support for robot picking, but in the actual picking environment, there are still problems such as inaccurate recognition, slow speed, and occlusion of ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T7/12G06K9/46G06K9/62
CPCG06T7/12G06T2207/20032G06T2207/20164G06T2207/30188G06V10/40G06V20/68G06V2201/07G06F18/253
Inventor 贾伟宽魏金梦张琦孙美丽赵艳娜丁艳辉
Owner SHANDONG NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products