Intelligent sensing machine blind guiding control method, system and device and storage medium

Abstract

The invention belongs to the technical field of computer vision, high-precision sensing and data fusion and artificial intelligence, and discloses an intelligent sensing machine blind guiding controlmethod, a storage medium, a system and a device. Through a online speech recognition library for recognition, the recognition results are sent to Baidu map API for destination selection, when the destination is selected, the equipment enters a standby mode, a sent external environment data packet is waited for; the current position information is sent to a mobile phone APP through a GSM module; course angle data is read, and preliminary processing is carried out; the key nodes in the path is marked; and a high-definition camera is controlled to acquire front road surface information in a multi-thread working mode, three pieces of information including a current moving direction, whether an obstacle exists in front and a current blind road position are calculated, and a front road conditionand the current position can be comprehensively judged. According to the invention, the social problems of long training time, high training cost, whether the blind guide dog is admitted to public places or not and the like of the traditional blind guide dog can be effectively solved.

Description

technical field [0001] The invention belongs to the technical fields of computer vision, high-precision sensing and data fusion, and artificial intelligence, and in particular relates to an intelligent perception machine guide blind control method, storage medium, system, and device. Background technique [0002] It is difficult to see the visually impaired travel in daily life, because the relevant barrier-free public facilities are not perfect, and the development of travel aids for the visually impaired is slow. At present, there are few ways to solve the life barriers of the visually impaired. The main method is to equip guide dogs. However, data shows that in 2016, there were 17.31 million registered visually impaired people in my country, but only more than 150 regular guide dogs were in service. Guide dogs are a powerful auxiliary means for the blind to travel, but their training costs are as high as 150,000 to 200,000 yuan, and the training process is difficult. Due...

AI Technical Summary

Problems solved by technology

[0005] At present, most of the common digital blind guide devices on the market are ranging devices using ultrasonic sensors and infrared sensors. They detect whether there are obstacles ahead through physical sensing. The functions are relatively single and cannot provide more comprehensive travel for the blind. auxiliary

At present, barrier-free facilities in most cities lack uniform standards, and there are situations such as malicious destruction and illegal occupancy. For example, parked bicycles are often seen on blind roads. The above situation greatly increases the difficulty for blind people to travel. It is difficult to use a single physical sensor Various obstacles in the complex environment of the outdoor street are detected, the user experience is poor, and there are certain dangers

[0006] With the development of deep learning technology and the improvement of the computing power of embedded computing platforms, it has become possible to use deep learning technology for target recognition on the mobile terminal, but how to integrate deep learning technology with traditional physical sensors to complete complex environments? There are still great difficulties in positioning, navigation, obstacle avoidance and other functions

Images