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

Airborne platform fast pedestrian detection system and method

A pedestrian detection and detection system technology, applied in the field of robot vision, can solve the problems of requiring dedicated hardware equipment, high-definition real-time image pedestrian detection and processing speed, etc., and achieve the effect of rapid deployment

Active Publication Date: 2016-11-23
SHANGHAI UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a fast pedestrian detection system and method on an airborne platform, which solves the problems of slow processing speed of pedestrian detection in existing high-definition real-time images or the need for dedicated hardware equipment

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
  • Airborne platform fast pedestrian detection system and method
  • Airborne platform fast pedestrian detection system and method
  • Airborne platform fast pedestrian detection system and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] see figure 1 and figure 2 , the airborne platform fast pedestrian detection system, including a high-definition camera (1), is characterized in that: the high-definition camera (1) is connected to an airborne fast processing module SECO CARMA DevKit (2); the high-definition camera (1) After capturing the high-definition image, the image is transmitted to the airborne fast processing module SECO CARMA DevKit (2) through the USB interface, and the fast processing module SECO CARMA DevKit (2) performs real-time pedestrian detection on the real-time collected high-definition image.

Embodiment 2

[0034] see figure 1 , image 3 and Figure 4 , the rapid pedestrian detection method on the airborne platform uses the above system for detection, and the detection steps are as follows:

[0035] Step 1: After the image is captured by the high-definition camera (1), it is input into the onboard fast processing module SECO CARMADevKit (2) in real time;

[0036] Step 2: The onboard fast processing module SECO CARMA DevKit (2) processes the input image as follows:

[0037] Color space standardization: The image read from the camera is directly transferred to the SECO CARMA DevKit embedded GPU using the transfer function of CUDA and processed, which is to read the value of each pixel of the data in the image and obtain the value of the pixel Then you can use the square root to perform Gamma correction, then proceed as follows: ,in Indicates the image pixel value after Gamma correction, Indicates the uncorrected value;

[0038] Calculation of gradient: The kernel funct...

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 an airborne platform fast pedestrian detection system and method. It includes a high-definition camera, which is connected to an airborne fast processing module SECO CARMA DevKit; after the high-definition camera captures high-definition images, the image is transmitted to the airborne fast processing module SECO CARMA DevKit through the USB interface, and real-time pedestrian processing is performed on the real-time collected high-definition images. detection. The method of the present invention is at first will calculate the eigenvector based on the gradient direction histogram of the pedestrian in the picture, comprise color space standardization, calculate the gradient, the gradient statistics of space and direction, the contrast standardization in overlapping blocks, eigenvector generation; Then based on The linear support vector machine classifies the feature vectors, and then judges that there are pedestrians in the picture according to the classification results. The embodiments of the present invention are mainly used for pedestrian detection calculations in images, especially pedestrian detection calculations in mobile robot embedded devices.

Description

technical field [0001] The invention discloses a fast pedestrian detection system and method on an airborne platform, and relates to the fields of robot vision, pattern recognition technology and CUDA parallel computing. Background technique [0002] Pedestrian detection is necessary for future intelligent vehicle assisted driving systems. It can effectively assist drivers driving vehicles in urban environments to respond to the external environment in a timely manner, avoid collisions with pedestrians, and reduce traffic accidents. It is due to the limited field of vision and fatigue of the driver at night. In addition, human body detection can also be used in the security real-time system of video surveillance to analyze and detect intrusion behavior from continuous video clips, so as to ensure the safety of personal and public personal property. [0003] Pedestrian detection is widely used in systems such as public security, video surveillance, intelligent assisted drivi...

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): G06K9/00G06K9/62H04N7/18
Inventor 刘恒利何旭栋黄潮炯颜春明李恒宇罗均谢少荣
Owner SHANGHAI UNIV