Quick visual positioning method for mobile robot
A mobile robot and visual positioning technology, applied in the field of intelligent robots, can solve the problems of increasing the required time and affecting the real-time performance of mobile robot visual positioning technology, so as to improve the efficiency of image processing, meet the real-time performance of visual positioning, and shorten the time required for image processing. effect of time
Inactive Publication Date: 2017-07-14
ZHEJIANG UNIV
View PDF5 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, with the development of CCD sensors and the improvement of self-positioning requirements, the resolution of images to be processed is getting higher and higher, and the time required for traditional image processing methods is greatly increased, which seriously affects the real-time performance of mobile robot visual positioning technology.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0064] Suppose the image to be processed is as image 3 As shown, the circle in the figure is the object to be extracted, and the image resolution is 1800*1200. The size feature is the diameter of the circle. The gray level threshold is set to 100 for image segmentation and the hough transform is used to extract the parameters of the circle. The diameter of the circle is about 194.7, and the minimum resolution size of the circle is 20. , the downsampling rate is about 100. According to formula (6), theoretically, the image processing workload of the method of the present invention is only 3.0% of that of the traditional method. using traditional methods for image 3 The time required for image processing and feature extraction is about 0.86s. Using the method in this paper to image 3 The time required for image processing and feature extraction is about 0.06s, and the actual time required is 7.0% of the traditional method, which is basically consistent with the theory. so...
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
Login to View More Abstract
The invention discloses a quick visual positioning method for a mobile robot. In the method, a CCD sensing module, an image processing module and a position information calculating module are mainly employed. A basic procedure of the algorithm includes the steps of: determining down-sampling rate on the basis of size characters and minimum resolution size of an object, performing down-sampling to an image, performing image processing to the down-sampled image to obtain coordinates of the object, determining the zone of the object in a source image according to the coordinates and the down-sampling rate of the object, and performing image segmentation and feature extraction to the zone in the source image. By means of the method, a high-resolution image is processed and converted into two low-resolution images, so that the work load and time of image processing are greatly reduced on the basis of ensuring visual positioning precision. The method satisfies a real-time demand of visual image processing well.
Description
technical field [0001] The invention relates to the field of intelligent robots, in particular to a fast visual positioning method for a mobile robot. Background technique [0002] Mobile robots are the carrier of artificial intelligence technology, and have broad application prospects in civilian fields such as civilian material handling, volcanic submarine exploration, intelligent cleaning, and military fields such as average reconnaissance, mine clearance, and anti-nuclear pollution. Self-localization is a basic requirement for mobile robots. Vision positioning technology is a main development direction of robot self-positioning technology. However, with the development of CCD sensors and the improvement of self-positioning requirements, the resolution of images to be processed is getting higher and higher, and the time required for traditional image processing methods is greatly increased, which seriously affects the real-time performance of mobile robot visual position...
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
Login to View More IPC IPC(8): G01C11/36G06T7/73
CPCG01C11/06G06T2207/10016G01C11/36
Inventor 车自远
Owner ZHEJIANG UNIV