Scene detection method and device, storage medium and terminal equipment

Abstract

The embodiment of the invention discloses a scene detection method and device, a storage medium and terminal equipment. The scene detection method comprises the steps: determining infrared feature information corresponding to a to-be-processed image and color feature information corresponding to the to-be-processed image in a weak gradient range based on the to-be-processed image; inputting the infrared feature information and the color feature information into a preset model, and outputting a prediction value corresponding to the to-be-processed image; and determining a detection result of the current scene based on the output prediction value.

Description

technical field [0001] The present application relates to the technical field of image processing, and in particular to a scene detection method, device, storage medium and terminal equipment. Background technique [0002] There are already many solutions for indoor / outdoor scene detection. For example, scene detection can be performed based on external devices (such as WIFI, light sensing, infrared, etc.), or based on the processing method of the image itself; here, the processing method based on the image itself can be divided into traditional threshold classification methods And classification methods based on machine learning (Marchine Learning). Among them, for indoor and outdoor scene detection, a good indoor and outdoor classification method can obtain better color reproduction. [0003] In the current indoor and outdoor classification methods, the extraction of features needs to rely on full-size images, and mobile terminals such as smart phones can usually only pr...

AI Technical Summary

Problems solved by technology

[0003] In the current indoor and outdoor classification methods, feature extraction needs to rely on full-size images, and mobile terminals such as smart phones can usually only provide reduced small-size images, so that the extraction method based on full-size images The accuracy of the feature is greatly reduced; in addition, the number of features extracted from the image is too large, which makes it difficult for the mobile terminal to process in real time, and the misjudgment rate is still low when performing indoor and outdoor scene detection based on the currently extracted features. relatively high

Images