Unlock instant, AI-driven research and patent intelligence for your innovation.

Infrared image pedestrian detection method and system based on deep learning

An infrared image and pedestrian detection technology, which is applied in neural learning methods, instruments, biological neural network models, etc. The effect of precision

Pending Publication Date: 2019-11-19
SHENZHEN BEIDOU COMM TECH CO
View PDF6 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As the most important and active factor in the environment, human targets have always been a research hotspot in the field of target tracking and detection. However, the non-rigidity of human targets and the shortcomings of infrared images make pedestrian detection based on infrared images full of difficulties. and challenge

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
  • Infrared image pedestrian detection method and system based on deep learning
  • Infrared image pedestrian detection method and system based on deep learning
  • Infrared image pedestrian detection method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] The present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments.

[0045] like figure 1 Shown, the inventive method has constructed FIDN (Fast-Infared-Detect-Network, fast infrared target detection) deep neural network, comprises the steps:

[0046] Step S1: Obtaining data and data preprocessing: Obtain infrared images containing pedestrians, preprocess the infrared images, and manually label the preprocessed infrared images, and then divide them into training sets and verification of detection models according to the set ratio set.

[0047]After obtaining a large number of pictures containing pedestrians, because the infrared images usually have poor imaging quality, some preprocessing is required, and then the processed infrared images are manually labeled. The label includes two parts, the target category and the target bounding box.

[0048] Step S2: Building a target detection FIDN network based on a c...

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 an infrared image pedestrian detection method and system based on deep learning, and belongs to the technical field of computer vision. The infrared image pedestrian detection method comprises the following steps: acquiring data and preprocessing the data; constructing a target detection FIDN network based on the convolutional neural network; constructing a target detectionFIDN network based on the convolutional neural network. Based on optimal model prediction, the invention also provides a detection system for realizing the infrared image pedestrian detection method.The beneficial effects of the invention are that the method can meet the real-time requirements while guaranteeing the high precision, and is high in robustness.

Description

Technical field [0001] The present invention relates to an image detection method, and in particular to an infrared image pedestrian detection method and detection system based on deep learning. Background technique [0002] Target detection is an important topic in the field of computer vision. The main task is to locate the target of interest from the image. It is necessary to accurately determine the specific category of each target and give the bounding box of each target. Target detection becomes a challenging task due to target deformation caused by factors such as viewing angle, occlusion, and posture. [0003] Traditional target detection methods are mainly divided into six steps: preprocessing, window sliding, feature extraction, feature selection, feature classification and post-processing. Traditional target detection generally designs some better artificial features and then uses a classifier for classification. As the requirements for target detection accuracy...

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): G06K9/00G06K9/62G06N3/08G06N3/04
CPCG06N3/08G06V20/40G06V20/52G06V2201/07G06N3/045G06F18/214
Inventor 孙立坤林保均王忠荣焦玉海吕建峰时文忠
Owner SHENZHEN BEIDOU COMM TECH CO