Method and device for scanning image effective area

Abstract

The invention provides a method for scanning an image effective area. The method comprises the steps that S1, an image to be searched is acquired; S2, an edge coordinate in the vertical direction of the image to be searched is obtained through a vertical searching algorithm; S3, an edge coordinate in the horizontal direction of the image to be searched is obtained through a horizontal searching algorithm; S4, an area defined by the line where the edge coordinate in the vertical direction is placed and the row where the edge coordinate in the horizontal direction is placed is determined as the image effective area. According to the method for scanning the image effective area, a pipeline processing mode is used, an embedded type environment is better achieved, the used vertical / horizontal searching algorithms have a filter effect, a low-pass effect on image edge noise is achieved, edge misjudgment caused by the noise is avoided, meanwhile, the method for scanning the image effective area can be directly used for Bayer RGB images, and the calculating complexity is lowered.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a scanning method and device for an effective image area in a sensor output image. Background technique [0002] When using an optical fingerprint sensor to collect fingerprint images, the existing image processing algorithm defaults that the effective image area is located in the middle of the collected image, and then processes the central area. However, due to the offset of the optical path, the effective area in the image acquired by CMOS has a corresponding At this time, the central area only contains part of the valid fingerprints, or even does not contain valid fingerprints at all. At the same time, noise in the invalid area will be introduced, which will eventually affect the effect of the subsequent algorithm. [0003] Optical fingerprint sensor output image has Figure 1 Descriptive features: Area 1 is the effective area of ​​the image; Areas 2 and 3 are caused by reflec...

AI Technical Summary

Problems solved by technology

For second-order operators such as Laplacian, due to its sensitivity to noise, the image needs to be smoothed before convolution, and the storage space of the image will be further increased

[0007] 2. Random heuristic search requires repeated accumulation of search results due to the random start point and end point, which requires a large amount of calculation and requires full image storage, which is not suitable for AISC implementation

This conversion also incurs additional computation

Images