Image binarization processing method, device, storage medium and electronic equipment

Abstract

Embodiments of the present invention provide an image binarization processing method, device, computer-readable storage medium, and electronic equipment. The image binarization processing method realized by hardware comprises: RAM initialization step: the RAM initialization that is used to record the number of pixels of each pixel value is zero; Image reading step: by direct memory access (DMA) unit, read Image to be processed; total product and value calculation step: through at least two groups of first multipliers, calculate in parallel the product and value of each of the pixel values ​​in its corresponding partition and its pixel value, and add each product and value Obtain the total product sum value GSUM; inter-class variance calculation step: through at least three second multipliers and dividers and according to the total product sum value GSUM, iteratively calculate the inter-class variance value G corresponding to each pixel value i , and get the current maximum between-class variance value G max ;Threshold determination step: the final maximum between-class variance value G max as the threshold for image binarization.

Description

technical field [0001] Embodiments of the present invention relate to image processing technologies, and in particular to an image binarization processing method, device, computer-readable storage medium and electronic equipment implemented by hardware. Background technique [0002] Digital image processing refers to the use of computers or other digital devices to process and process image information. This technology is developing extremely rapidly and has a wide range of applications. [0003] Image processing (or processing, analysis) is mainly divided into three aspects: [0004] (a) Improving the visual quality of the image, such as performing image brightness and color transformation, enhancing and suppressing certain components, and performing geometric transformation on the image, etc., to improve the quality of the image. [0005] (b) Extract some features or special information contained in the image. The extracted features may include, for example, frequency do...

AI Technical Summary

Problems solved by technology

During the processing, it is necessary to perform read and write operations on the image memory, and each read and write takes dozens of processor cycles, which seriously slows down the efficiency

[0009] In addition, the Otsu algorithm implemented by the software adopts a sequential calculation method during the execution process, that is, operations such as multiplication and division are performed sequentially, which further increases the processing time

[0010] Due to the low efficiency of software algorithms, for time-limited application scenarios, it is necessary to use high-frequency processors, thereby increasing the overall power consumption of the system

At the same time, with the improvement of image quality and application requirements, software algorithm implementation will be further inefficient, making its algorithm unable to adapt to more application scenarios

Images