Method for detecting maturity of fruits on tree based on visual saliency map

A detection method and maturity technology, applied in image enhancement, image analysis, measuring devices, etc., to achieve the effects of slowing down randomness, compensating for strong randomness, and strong learning ability

Pending Publication Date: 2022-05-27
SOUTH CHINA AGRI UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the shortcomings in the prior art, and provide a method for detecting fruit maturity on trees based on visual saliency maps, which has good accuracy and robustness, and is suitable for complex scenes and light conditions in natural environments. changeable situation

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
  • Method for detecting maturity of fruits on tree based on visual saliency map
  • Method for detecting maturity of fruits on tree based on visual saliency map
  • Method for detecting maturity of fruits on tree based on visual saliency map

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] This embodiment is suitable for detecting the ripeness of fruits on trees in complex natural environment, wherein the complexity of natural environment is mainly reflected in the phenomenon of inconsistent light conditions (including front light, side light, backlight, etc.) and fruit maturity. Fruits at different maturity stages may coexist on the same tree. This embodiment takes the ripeness detection of citrus on trees as an example, and proposes a robust and robust specific measures for fruit ripeness detection oriented to complex natural environment. like figure 1 As shown, the method for detecting ripeness of fruits on trees based on visual saliency map includes the following steps:

[0046] (1) Use camera equipment to take citrus images at different distances and under different lighting conditions within a distance of 0.5m to 1.5m. In order to speed up the processing speed of the algorithm, the image size is uniformly reduced to the original Figure 4 one pa...

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 discloses a method for detecting maturity of fruits on a tree based on a visual saliency map. The method comprises the following steps: firstly, acquiring a fruit tree image through a camera, and identifying the fruits on the tree by using a YOLOv5 target detection algorithm; cutting fruit image blocks by using a bounding box coordinate output by the YOLOv5, and obtaining a visual saliency image by using an improved MSSS visual saliency detection algorithm; and the fruit image blocks and the corresponding visual saliency images are connected in series to form a four-dimensional RGBS image, and the maturity category of the fruit is judged by using an image classification network ResNet34 in combination with RGB and saliency information of the fruit local image. The detection method provided by the invention is excellent in accuracy and robustness, and is suitable for the conditions of complex scene and changeable illumination in a natural environment.

Description

technical field [0001] The invention belongs to the field of intelligent picking, and in particular relates to a method for detecting ripeness of fruits on trees based on a visual saliency map. Background technique [0002] Fruits on trees have different maturity in the natural environment, and it is of great significance to realize the selective harvesting of fruit picking robots. If the picking robot can identify the three growth stages of immature, semi-ripe and mature fruit on the tree, the subsequent tasks of handling, warehousing, grading and sales can be arranged more reasonably, which is crucial for intelligent picking operations. There are two main types of fruit maturity detection methods: one is to manually extract color features and use machine learning classifiers to build a maturity discrimination model; although the manual features are deterministic, the trained classifier is easy to use in natural environment. Failure, generalization is weak. The other is t...

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): G06T7/90G06T7/11G06T5/00G06N3/04G06K9/62G06V10/46G06V10/774G06V10/764G01N21/84
CPCG06T7/11G06T7/90G06T5/002G01N21/84G06N3/045G06F18/2411G06F18/214
Inventor 熊俊涛陈淑绵焦镜棉谢志明霍钊威胡文馨韩咏林熊春源
Owner SOUTH CHINA AGRI UNIV
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