Image visual effect enhancement method based on image tonality
An image vision and image technology, applied in the field of image processing, can solve the problems that the algorithm model cannot well adapt to the brightness requirements, the photo exposure is unnatural, the image is too bright, etc.
Active Publication Date: 2019-09-13
重庆米弘科技有限公司
View PDF12 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Therefore, although the above methods can also enhance the visual effect of the image, they all have a disadvantage, that is, the algorithm model cannot well adapt to the brightness requirements of each image.
Through the evaluation of professional retouchers and the statistical evaluation of experimental results, if all the tonality images are mixed together for training, it is found that the image processing of the middle tonality has a very good effect, and the high-key and low-key images are common. There are cases where the processing effect is poor. For example, for high-key photos, the image will become too bright; for low-key photos, the photos will be exposed unnaturally; some photos may even appear color cast
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0054] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.
[0055] The invention is used for image data preprocessing, and classifies the images according to the exposure degree of the images themselves. The present invention divides images into three categories according to the principle of global exposure and the principle of partial exposure: high-key images, mid-key images and low-key images. After classification, the original image and target image pairs of different tonality categories are input into their respective CAN networks for learning, and three different models (with different model parameters) are obtained. For the network architecture, please refer to the attached figure 2 . During the test, the tonality of the test image should also be judged to ensure that the...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to an image visual effect enhancement method based on image tonality, and the method comprises the steps: S1, collecting a super-definition image, and enabling all collected high-definition images to form an original image set; S2, according to global and local exposure principles, dividing the original images into a high-speed image, a medium-speed image and a low-speed image; S3, performing post-processing on the original image to enhance the visual effect, and obtaining a processed high-speed image set; S4, taking an image sequence formed by the original image with thesame adjustability and the corresponding post-processed image as input, and training the network to finally obtain three models; and S5, judging the tonality of the to-be-processed image, and then inputting the to-be-processed image into the model with the same tonality as the to-be-processed image for processing to obtain an image with an enhanced visual effect. The method is simple and effective, the experiment result shows that after the images are classified based on the adjustability, model processing is established, and the effect of the obtained result image is better.
Description
technical field [0001] The invention relates to the technical field of image processing, in particular to a method for enhancing image visual effects based on image tonality. Background technique [0002] With the continuous development of economy and society, the continuous advancement of science and technology, and the continuous reduction of the cost of image acquisition and storage, using cameras to record beautiful moments in life has become an important way for people to preserve memories. Digital image and its processing have gradually penetrated into every corner of our life and work. Image enhancement is an important branch of digital image processing, and its purpose is mainly to improve the visual effect of the image for a given image application. [0003] In recent years, many people have applied deep learning to the image field and achieved a lot of success in image enhancement. For example, TaoL, Zhu C, and Xiang G proposed a low-brightness image enhancement me...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06T5/00G06T5/50
CPCG06T5/00G06T5/50G06T2207/20081G06T2207/20084G06T2207/10016Y02D10/00
Inventor 杨梦宁徐忠涂波宇李小斌
Owner 重庆米弘科技有限公司