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

A detection method for leaf occlusion based on deep learning and traditional algorithms

A detection method and deep learning technology, applied in the field of image processing, can solve problems such as the impact on public safety and missing monitoring scenes, and achieve the effects of reducing the impact on public safety, preventing missing, and improving accuracy

Active Publication Date: 2021-04-06
ANHUI SUN CREATE ELECTRONICS
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If the surveillance camera blocked by the leaves cannot be detected, the surveillance scene will be missing, which will affect public safety

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
  • A detection method for leaf occlusion based on deep learning and traditional algorithms
  • A detection method for leaf occlusion based on deep learning and traditional algorithms
  • A detection method for leaf occlusion based on deep learning and traditional algorithms

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] A method for detecting leaf occlusion based on deep learning and traditional algorithms, comprising the following steps:

[0038] S1, adjust the size of each test set image and each training set image collected so that all images have the same size, and the size of all images in this embodiment is 64*64; for each test set image and each The training set images are rotated to the left and right, and each test set image and each training set image can derive two images, which increases the sample size; adjust the contrast of each test set image and each training set image, Make the same image have multiple contrasts, increase the sample size, and adjust the brightness of each test set image and each training set image to increase the sample size of the test set image and training set image.

[0039] Specific steps are as follows:

[0040] S10, adjusting the size of each collected image so that all the collected images have the same size, and the resized image is set as t...

Embodiment 2

[0056] On the basis of embodiment 1, table 1 is the test set image, sample number unit: Zhang

[0057] Table 1

[0058]

[0059] In Table 1: the image without leaf occlusion is an image without leaf occlusion including urban and rural multi-scenes;

[0060] Detection rate = the number of images detected to be occluded by leaves / the number of images actually occluded by leaves

[0061] Accuracy rate = the number of images detected to be occluded by leaves and actually occluded by leaves / the number of images detected to be occluded by leaves.

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 relates to the technical field of image processing, in particular to a method for detecting leaf occlusion based on deep learning and traditional algorithms. Input the training set into the convolutional neural network model, train the convolutional neural network model until the loss function of the convolutional neural network model meets the set conditions, terminate the training of the convolutional neural network model, and the convolutional neural network model after training The product neural network model is the best classification model. The traditional algorithm is used to judge whether the test set image is an image without leaf occlusion. If the traditional algorithm detects that the image is occluded by leaves, a deep learning algorithm is used to perform a one-step detection on the image. It can detect whether the surveillance camera is blocked by leaves, which is convenient for staff to know in time that the surveillance camera is blocked by leaves, so as to adjust the position of the surveillance camera, prevent the lack of surveillance scenes, and reduce the impact on public safety.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for detecting leaf occlusion based on deep learning and traditional algorithms. Background technique [0002] With the rapid growth of video surveillance business, its existing problems are gradually exposed. How to find the fault of the front-end surveillance camera in the first time and improve the operation and maintenance efficiency of the video surveillance system is an indispensable part of the development of the video surveillance system. [0003] Leaf occlusion in surveillance camera faults is one of the most common faults. Since the position of the surveillance camera remains unchanged, the trees near it will continue to grow over time. In a specific season or time, the originally normally installed surveillance camera will be blocked by leaves. . If the surveillance camera blocked by the leaves cannot be detected, the surveillance scene will be missin...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08G06Q50/26
CPCG06N3/08G06Q50/26G06V20/10G06N3/045G06F18/241G06F18/214
Inventor 唐艳艳王一灵徐金凤李贤军薛家彬刘升谢永亮王梦园吉江燕范联伟余保华
Owner ANHUI SUN CREATE ELECTRONICS
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com