Adaptive height modified histogram equalization and chroma correction in YCbCr color space for fast backlight image compensation
Our Price
$33.80
10000 in stock
Support
Ready to Ship
Description
The color correction and enhancement of the videos have many applications mainly in cameras for producing natural effects in the images. For color correction always the intensity all over the video frames is analyzed and the observed intensity is applied all over the image. A method that uses the intensity of the face image for the color correction all over the image is proposed. From the detected face images the illumination of the skin is detected. The skin illumination is not obtained for the patch means a state of art method that is based on thresholding is used to apply color correction. The performance of the process is measured by the calculation of PSNR, MSE and detection rate. The input video is converted into frames. The face region is detected from the frames. The face region is alone mapped. The mapped face regions were converted from RGB color space to YcBr color space. The histogram of the images were compensated based on threshold. The image histograms higher than the particular threshold is converted into a particular value inorder to enhance the image and to equalize the histogram all over the image. The modified histogram is mapped into the original frames and the mapped image is obtained. The mapping parameters were estimated and based on the estimated parameters the mapping process is employed. The Low Contrast Map1 and Low Contrast Map2 were generated that were helpful in the identification of the pixels that were needed to be enhanced. The final Low contrast map is obtained and extracted Low contrast map is applied to the Ycbr color space. The color space is transformed based on the obtained Low contrast map. The resulting images were converted from Ycbr color space to RGB color space. The regions other then the face region is replaced from the original frames. The performance of the process is measured with the help of PSNR, MSE and Detection rate.