Image Detail Enhancement Method

Abstract

The invention provides an image detail enhancement method. The method comprises the following steps: acquiring a plurality of original images with different exposure degrees of the same scene; calculating gradient fields of a plurality of original images respectively; calculating a fusion gradient field of the scene according to a gradient field of a plurality of original images; reconstructing the details of the scene according to the fusion gradient field of the scene to enhance the image. The method provided by the invention can fully utilize the information of the image to be fused, so that the fused image simultaneously comprises dark field information and bright field information of a plurality of original images with different exposure degrees, thereby expanding the dynamic range ofthe image and better reflecting the detail information of the image. Compared with the images obtained according to other image detail enhancement methods, the images obtained according to the methodof the present invention retain rich scene detail information, which is helpful for target identification or further processing, and has better real-time performance.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an image detail enhancement method. Background technique [0002] When shooting and displaying images, since the dynamic range captured by most of the current shooting equipment is limited, which is far smaller than the dynamic range of the real scene, it is prone to local overexposure or overdarkening when actually shooting images, which will lead to details in the image. Information is lost. In this case, although the detailed information within a certain dynamic range can be obtained by adjusting the parameters of the shooting equipment, the detailed information outside the range is still unrecognizable. [0003] In response to the above situation, there are currently a variety of image detail enhancement technologies, which are generally divided into single-frame technology and multi-frame technology: single-frame-based detail enhancement methods include automatic gain contro...

AI Technical Summary

Problems solved by technology

[0002] When shooting and displaying images, since the dynamic range captured by most of the current shooting equipment is limited, which is far smaller than the dynamic range of the real scene, it is prone to local overexposure or overdarkening when actually shooting images, which will lead to details in the image. information loss

In this case, although the detailed information within a certain dynamic range can be obtained by adjusting the parameters of the shooting equipment, the detailed information outside this range is still unrecognizable

[0003] In response to the above situation, there are currently a variety of image detail enhancement technologies, which are generally divided into single-frame technology and multi-frame technology: single-frame-based detail enhancement methods include automatic gain control, histogram method, and multi-scale image enhancement technology Such as Retinex and other methods, these methods are prone to loss of detail information when processing images with a large dynamic range

The method based on multi-frame technology includes the fusion method based on the evaluation factor and the method based on the camera response curve. Although the fusion method based on the evaluation factor is simple, it is easily affected by the performance of the evaluation factor in actual processing; while the method based on the camera response curve Although the method has a better detail enhancement effect, its real-time performance is poor, which is not conducive to practical engineering applications.

Images