Fruit recognition tracking method and system based on deep learning algorithm

A deep learning and fruit recognition technology, applied in the field of recognition and tracking, can solve the problems of easy loss of targets, low accuracy of target recognition, poor target positioning accuracy, etc.

Active Publication Date: 2018-09-21
北京禾泽方圆智能科技有限公司
View PDF6 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Existing fruit identification and tracking technologies include the use of binocular vision methods to identify and track target fruits, and the recognition and positioning of fruits through convolutional neural networks and support vector machines, but the identification and tracking process that exists in the prior art still has the following defects: The existing target recognition accuracy is low, the target positioning accuracy is poor, and the tracked target is easy to lose

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
  • Fruit recognition tracking method and system based on deep learning algorithm
  • Fruit recognition tracking method and system based on deep learning algorithm
  • Fruit recognition tracking method and system based on deep learning algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] Embodiment 1 of the present invention provides a method for identifying and tracking fruits based on deep learning algorithms, such as figure 1 As shown, the method includes:

[0054] S1: Acquire the image of the pre-tracking part;

[0055] S2: Use the deep learning algorithm to identify the target in the image, and obtain the position of the target in the image;

[0056] S3: Track the location of the target.

[0057] needs to be specified, such as figure 2 As shown, step S2 specifically includes:

[0058] S21: Use the deep recognition network algorithm to identify target features, and calculate the confidence of each target feature;

[0059] S22: Compare the confidence of each target feature with a confidence threshold, and when there is a target feature greater than the confidence threshold, proceed to step S23;

[0060] S23: Determine the number N corresponding to the target feature greater than the confidence threshold, if N=1, go to step S24, if N>1, go to st...

Embodiment 2

[0069] Embodiment 2 of the present invention provides a fruit recognition and tracking method based on a deep learning algorithm, which is basically the same as that of Embodiment 1, the difference is that the preset number of frames is 100 frames, and by setting the preset number of frames to 100 frame, which can ensure the timely recognition and tracking of the next frame of images, and ensure the efficiency of recognition and tracking. If the number of frames is too small, there will be recognition and tracking process interleaving, which will cause errors in recognition and tracking. If the number of frames is too large, there will be recognition and tracking errors. Tracking is discontinuous, reducing work efficiency.

Embodiment 3

[0071] Embodiment 3 of the present invention provides a method for identifying and tracking fruits based on a deep learning algorithm, which is basically the same as that in Embodiment 2. The difference is that, as Figure 4 As shown, step S3 also includes:

[0072] S34: When step S32 judges that the target is not lost, judge the time t tracked by the depth tracking network algorithm and compare it with the time threshold t 1 Make a judgment, when t1 , proceed to step S35, when t≥t 1 , proceed to step S2;

[0073] S35: sending a picking instruction to the robot;

[0074] S36: When step S32 determines that the target is lost, at the same time send an instruction to the robot to stop moving until step S35 is performed.

[0075] In the identification tracking method provided by the present invention, the tracking time is further judged, and when the tracking time reaches a preset threshold, step S2 is performed again. This improves the overall tracking efficiency; when the ta...

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 provides a fruit recognition tracking method and system based on a deep learning algorithm. The method comprises the following steps: obtaining the image of a pre-tracking part, using the deep learning algorithm to recognize a target in the image, and obtaining the position of the target in the image; and tracking the position of the target. According to the method and the system provided by the invention, the recognition and tracking of fruits in the image are realized in combination with a deep recognition network algorithm and a deep tracking network algorithm, the accuracy oftarget recognition and the correctness of tracking under complicated natural environments are guaranteed, the tracked target is unlikely to lose, accurate picking positioning is provided for pickingoperation, and the efficiency of picking operation is improved.

Description

technical field [0001] The invention belongs to the technical field of identification and tracking, and in particular relates to a method and system for identification and tracking of fruits based on a deep learning algorithm. Background technique [0002] With the development of computer technology and information collection and processing technology, robots have gradually entered the field of agricultural production. In recent years, picking robots for automatic fruit harvesting have attracted more and more attention. The picking robot identifies the target fruit from the fruit growth environment, then tracks the target fruit to obtain its spatial position, and finally uses the picking execution part to pick the target fruit to complete the fruit automatic harvesting operation. Since the picking execution part relies on the target fruit recognition and tracking technology to provide Therefore, the fruit identification and tracking technology plays an important role in ensu...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/246G06T7/11G06T7/136G06T7/194G06T7/73
CPCG06T2207/10016G06T2207/20081G06T7/11G06T7/136G06T7/194G06T7/246G06T7/73
Inventor 毕松吴劲松陈俊文张潞高峰刁奇
Owner 北京禾泽方圆智能科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap