Signal lamp foreground area identification method and related device

Abstract

The invention discloses a signal lamp foreground area identification method. The method comprises the following steps: determining a candidate area from a signal lamp image; performing scene recognition on the candidate area by adopting a classification network to obtain a corresponding scene type; and determining a foreground region extraction operation according to the scene type, and executingthe foreground region extraction operation on the candidate region to obtain a foreground region. Through scene identification of the signal lamp image, a foreground region extraction operation corresponding to the scene type can be selected, the identification accuracy and precision are improved, and the efficiency of the identification process is improved. The invention also discloses a signal lamp foreground area identification device, a server and a computer readable storage medium, which have the above beneficial effects.

Description

technical field [0001] The present application relates to the field of image technology processing, and in particular to a method for identifying a foreground area of ​​a signal light, a device for identifying a foreground area of ​​a signal light, a server, and a computer-readable storage medium. Background technique [0002] With the continuous development of image processing technology, image acquisition devices or monitoring devices are applied in various environments in order to perform corresponding processing on acquired images. At present, in order to accurately capture vehicle violations in urban traffic, lane monitoring cameras are used in urban traffic arteries. When vehicle violations occur, capture the corresponding violation record map in time. However, illegal records are easily affected by the camera's own imaging, and the color of the signal light appears color cast, for example, the red light is yellowish and white. Reduced the persuasiveness of the evide...

AI Technical Summary

Problems solved by technology

However, when the application is applied in a more complex environment, not only the shooting scenes are varied, but also the gap between each shooting scene is large, and the same extraction method will make the extraction accuracy rate low and reduce the execution efficiency of the extraction process.

Images