3results about How to "High sharpness" patented technology

The application belongs to the field of image vision technology, and discloses a sand-dust image sharpening method, system, medium, equipment and terminal. The improved automatic color scale algorithm is used to perform secondary distribution on image pixels, so as to reduce image color deviation. The gamma correction algorithm and the improved dark channel prior algorithm are combined to perform brightness compensation on each image pixel, so as to enhance image sharpness. The gray world algorithm is combined to balance image color, so as to realize sharpening of deep-concentration sand-dust images. The application can effectively improve sand-dust image quality, improve the definition and sharpness of degraded images, and realize sharpening of blurred images in sand-dust weather. The application can effectively improve the definition and sharpness of deep-degraded sand-dust images. The image collected in severe weather with large sand-dust can be restored to a clear image by the processing of the application, which has important practical significance for security monitoring, road monitoring, face recognition and the like.

This application provides a low beam headlight module and a vehicle headlight. The low beam headlight module includes: multiple light sources arranged at intervals along a first direction; the low beam lens includes a widening lens unit and an elbow lens unit, which are integrally formed structures. The widening lens unit and the elbow lens unit share a light-incident surface and a light-exit surface. The light-incident surface is arranged opposite to the multiple light sources along a second direction, and the light-exit surface is located on one side of the light-incident surface along a third direction; the widening lens unit is used to form a low beam widening light pattern, and the elbow lens unit is used to form a low beam elbow cutoff line light pattern. The low beam elbow cutoff line light pattern and the low beam widening light pattern are combined to form a low beam light pattern; the light-exit surface includes multiple sub-light-exit surfaces arranged sequentially along the first direction, and the distance between the root of each sub-light-exit surface and the light source in the third direction gradually increases along the first direction. The embodiments of this application can effectively improve the luminous efficacy of the low beam headlight module.

This application discloses an image processing method, apparatus, device, and readable storage medium, relating to the field of image processing technology. The method includes: acquiring a first image; determining frequency domain features corresponding to the first image; fusing the first image and the frequency domain features to obtain a second image; determining noise features in the first image based on the second image and the first image; and removing noise from the second image based on the noise features to obtain a third image. This method can improve the denoising effect of images.