A bpnn-based fast gamma correction method and device

Abstract

The invention discloses a fast Gamma correction method and device based on BPNN, which includes training the BPNN model to obtain an initial value prediction model, using the initial value prediction model to predict the initial value of the register, determining the register value correction model, and predicting the RGB register correction through the model value steps; this data-based Gamma correction method and device trains a model through a back-propagation neural network, makes predictions based on the trained model, collects data from a small number of test samples, and inputs it into the trained BPNN model. The BPNN model determines the characteristics of the test sample based on these data. This characteristic is different from the test sample and can be related to the training sample, making the determination of the initial value and precise value of the register simple and efficient, and making accurate predictions; its trained The model has autonomous learning and prediction capabilities, which effectively overcomes the difficulty of obtaining initial values ​​with traditional algorithms and the long time-consuming correction of approximation algorithms. It can quickly perform Gamma correction on panels, improving accuracy and efficiency.

Description

Technical field [0001] The present invention belongs to the technical field of machine learning and image processing, and more particularly, to a fast BPNN based Gamma correction method and apparatus. Background technique [0002] Gamma correction is a technique widely used in photography, video images and computer graphics; Gamma refers to the non-linearity between the electrical signal and the optical characterization of luminance gradation, generally a non-linear device such intrinsic photoelectric characteristics of the decision. In manufacturing, the variation due to different processes and materials used, the drive voltage of the OLED device Gamma curves differ, Gamma correction to bring considerable difficulty. [0003] Gamma correction refers to the intermediate gray values ​​to match changes Gamma monitor; reference figure 1 , Conventional long Gamma correction process includes a start correction, acquires RGB register initial value, using exact values ​​for the register...

AI Technical Summary

Problems solved by technology

[0004] For the above defects or improvement needs of the prior art, the present invention provides a fast Gamma correction method and device based on BPNN (BackPropagation Neural Net, reverse propagation neural network), the purpose of which is to solve the difficulty of obtaining the initial value of the register in the prior art , Determining the precise value of the register takes a long time

Images