Safety belt detection method and device, electronic equipment and storage medium

Abstract

The invention discloses a safety belt detection method and device, electronic equipment and a storage medium, relates to the fields of artificial intelligence, deep learning and image detection, and can be applied to the field of automatic driving. According to the specific scheme, a to-be-recognized image is input into a pre-trained semantic segmentation network, safety belt area detection is conducted on the to-be-recognized image through the semantic segmentation network, and a safety belt detection area of the to-be-recognized image is obtained; and inputting the safety belt detection areainto a pre-trained safety belt detection network, and performing safety belt detection on the safety belt detection area through the safety belt detection network to obtain a safety belt detection result corresponding to the safety belt detection area. According to the embodiment of the invention, whether the driver fastens the safety belt or not can be accurately detected, so that a vehicle supervision department can be effectively assisted in operation supervision, and a guarantee is provided for safe driving of vehicles.

Description

technical field [0001] The present application relates to the field of computer technology, and further relates to the fields of artificial intelligence, deep learning and image detection, which can be applied to the field of automatic driving, especially a safety belt detection method, device, electronic equipment and storage medium. Background technique [0002] With the continuous development of the Internet and artificial intelligence technology, more and more fields have begun to involve automatic computing and analysis, among which the field of surveillance and security is one of the most important scenarios. [0003] For public operating vehicles, such as taxis, buses, long-distance buses, etc., the driver's driving safety is particularly important due to the safety of many passengers. Therefore, many public operating vehicles have been installed with on-board monitoring cameras to facilitate the corresponding companies or regulatory authorities to monitor the driver'...

AI Technical Summary

Problems solved by technology

In the existing schemes, the artificial naked eye method has disadvantages such as slow speed, large error, and high time and labor costs; the direct classification method based on CNN, because the safety belt target is small in the image, the features that can be extracted are scarce, and the surrounding There is also a large amount of interference information, which leads to low recognition accuracy and unsatisfactory recognition effect in real vehicle scenes

Images