A terrain detection method for blind guide sticks based on active vision guidance
A technology of terrain detection and active vision, applied in the direction of devices to help people walk, physical therapy, etc., can solve problems such as not much progress
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Publication Date
- 2020-12-11
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The invention relates to a terrain detection method for blind guide sticks based on active visual guidance, and belongs to the technical field of active guidance and deep learning. Background technique
[0002] In recent years, with the popularity of electronic products and changes in people's lifestyles, the number of people with visual impairment or blindness around the world has been increasing. According to the World Health Organization report in 2019, at least 2.2 billion people worldwide are visually impaired or blind. What is even more worrying is that the number of blind people in the world is increasing every year, and blind people have become a group that cannot be ignored in the world.
[0003] As one of the five sensory channels of human beings, the sense of touch is the basic channel for information interaction between humans and the outside world. Through the tactile channel, important information such as hardness, temperature, shape, an...
Examples
Embodiment Construction
[0036] The terrain detection method of the blind guide stick based on active vision guidance proposed by the present invention, its flow chart is as follows figure 1 shown, including the following steps:
[0037] (1) Obtain ground images with the camera 1 on the glasses for the blind, such as figure 2 shown;
[0038] (2) the ground image is processed by the integrated chip 2 on the glasses for the blind, and the ground image of step (1) is segmented to obtain a plurality of image blocks to adopting the uniform block method;
[0039] In this step, the image is segmented using a block method, and commonly used block methods include uniform block method, superpixel segmentation method, etc. Among them, the size of local blocks after superpixel segmentation is different, and it cannot guarantee that the proportion of image information occupied by each feature point is consistent. In order to avoid introducing more interference parameters, the uniform block method is used to se...