Auxiliary obstacle perception method for visually impaired people based on improved YOLO model

Abstract

The invention discloses an auxiliary obstacle perception method for visually impaired people based on an improved YOLO model, and the method comprises the steps: employing Darknet-YOLOv3 as a framework, and employing Darknet-53 as a feature extraction backbone network; according to the YOLOV3 algorithm, carrying out feature fusion by using a feature map up-sampling idea in a feature pyramid network FPN, so that the precision of small target detection is improved, and various common obstacles, including road cones, stone balls, isolation columns, forbidden cross bars, handrails, fire hydrants, plants, people, pits, water pits and the like, on sidewalks can be detected and identified. The method can identify various identifications and targets at traffic intersections, including zebra crossings, signal lamps, bicycles, motorcycles, vehicles, people and the like, and can also judge upstairs, downstairs, various steps and some other obstacle targets of unknown types.

Description

technical field [0001] The invention relates to the field of blind guides, in particular to an improved YOLO model-based auxiliary obstacle perception method for visually impaired persons. Background technique [0002] China is the country with the largest number of blind people in the world, about 12 million, accounting for 18% of the world's blind population. As a special group of social groups, they live in boundless darkness all their lives, so they often encounter various problems. The current market Most of the guide products on the Internet are simple in structure and single in function (it can only simply remind you that there are obstacles ahead). Although some products are easy to use, the auxiliary effect is not obvious. Moreover, blind friends will encounter many problems when using them, such as The road conditions are bad, potholes are uneven, there are obstacles hanging ahead, etc. Ordinary blind guide products cannot be accurately detected. The obstacle dete...

AI Technical Summary

Problems solved by technology

[0002] China is the country with the largest number of blind people in the world, about 12 million, accounting for 18% of the world's blind population. As a special group of social groups, they live in boundless Therefore, they often encounter various problems. Most of the guide products currently on the market are simple in structure and single in function (it can only simply prompt that there are obstacles ahead). Although some products are easy to use, the auxiliary effect is not obvious. , Moreover, blind friends will encounter many problems when using them, such as poor road conditions, uneven potholes, hanging obstacles in front, etc. Ordinary blind guide products cannot accurately detect

The obstacle detection function of the existing blind guide products is only limited to the detection of the distance of the obstacle, and cannot accurately locate the position of the obstacle, and can only detect a single obstacle, such as the detection of multi-moving obstacles. Among them, only the obstacle closest to the user can be detected, so the practicability of the product's blind-guiding function is greatly reduced

[0003] There are international and domestic teams and companies who have conducted research on smart blind-guiding equipment such as guide glasses, but due to unsatisfactory performance and user experience, etc. In the stage of performance testing and small batch trial production, no large-scale market has been formed so far

Especially in China, the research and development of auxiliary blind-guiding equipment for the blind is still in its infancy, and there is still a long way to go before large-scale productization and commercialization. Moreover, at present, the market space for blind-guiding equipment in my country is still limited. Far from reaching the stage of development and popularization, it is particularly important to solve such problems

Images