Multi-modal target detection method and device, sweeper and storage medium

Abstract

The invention discloses a multi-modal target detection method and device, a sweeper and a storage medium, and the method comprises the steps: obtaining an environment image corresponding to the current position of the sweeper, extracting the feature information corresponding to a target region in the environment image, so as to recognize the environment where the sweeper is located, and obtaining an environment recognition result, the environment recognition result comprises a bright environment and a dark environment, if the environment recognition result is the dark environment, obtaining a target to-be-detected image corresponding to the dark environment, and based on the target to-be-detected image, performing target detection through a multi-modal target detection model to obtain a target detection result, the multi-modal target detection model is obtained by performing iterative training based on pre-obtained to-be-trained images of different modals. The technical problem that the sweeping effect of the sweeper is poor due to the fact that the sweeper is low in target recognition degree in the night environment and cannot accurately avoid obstacles in the sweeping process is solved.

Description

technical field [0001] The present application relates to the technical field of sweepers, and in particular, to a multimodal target detection method, device, sweeper and storage medium. Background technique [0002] With the rapid development of smart homes, more and more smart home appliances have entered people's lives. In order to free people's hands and reduce the time occupied in housework, sweepers are widely used in people's lives. At present, sweepers are cleaning During operation, the camera is usually used for real-time shooting to identify objects. However, the camera of the sweeper can often only work in bright scenes during the day. In dark scenes, the pictures collected by the camera are very dark, and you can't even see anything. As a result, the target recognition degree of the sweeper in the night environment is low, and the obstacle cannot be accurately avoided during the cleaning process, resulting in poor cleaning effect of the sweeper. SUMMARY OF THE ...

AI Technical Summary

Problems solved by technology

[0003] The main purpose of this application is to provide a multi-modal target detection method, device, sweeper and storage medium, aiming to solve the problem that the sweeper in the prior art has a low target recognition degree in the night environment and cannot be accurately avoided during the cleaning process. Open obstacles, leading to technical problems of poor cleaning effect of the sweeper

Images