Full-reference image quality assessment method based on image saliency detection
An image quality evaluation and full reference technology, applied in image analysis, image enhancement, image data processing, etc., can solve the problems of inaccurate and repeatable evaluation, poor operability, slow speed, etc., and achieve ease of dimensionality The problem, the idea is simple, and the effect of the simplified algorithm
Active Publication Date: 2022-02-01
XIDIAN UNIV
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The "Mean Opinion Score" (MOS, Mean Opinion Score), which uses a large number of observers to score and take the average, has been the optimal strategy for evaluating digital images for a long time, but some defects of this method cannot be ignored, because The observation motivation, knowledge background, observation environment and psychological state of the observers are all different, so it is impossible to make accurate and repeatable evaluations. Secondly, this method needs to spend a lot of manpower and material resources, and the speed is slow and the cost is too high. Not strong, and this method cannot be described by a mathematical model, it is difficult to be widely used in industrial production
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
[0080] Objective image quality evaluation methods can be divided into: full reference, no reference and partial reference image quality evaluation methods. The invention belongs to a full-reference image quality evaluation method.
[0081] Robust visual saliency enables proper processing of images without prior knowledge. Through the study, it was thought that human cortical cells might be hard-coded in their receptive fields to preferentially respond to high-contrast stimuli. Detection methods based on global contrast tend to separate large-scale objects from their surroundings. This method outperforms those local contrast methods that usually yield higher saliency only near contours. Global considerations can assign similar saliency values to similar regions in an image, and can evenly highlight objects. The salience of a region is mainly determined by its contrast with the surrounding regions, and regions that are far away play a smaller role.
[0082] The present inv...
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 discloses a full-reference image quality evaluation method based on image saliency detection, which includes the following steps: space conversion (extracting a distorted image and its corresponding original image, converting the image from RGB color space to Lab color space), Image segmentation (segmentation of color images in Lab color space into regions), RCC saliency detection, superpixel segmentation (superpixel segmentation of distorted images in RGB color space and their corresponding original images), AMC saliency detection, VSI calculations. The benefits of the present invention are: (1) We obtain the salient region of the image through saliency detection, and use the extracted saliency image to evaluate the image quality, bypassing the problem of difficult modeling of the human visual system, so the present invention The evaluation method provided by the invention is simple in thinking, better conforms to the characteristics of the human visual system, and has better consistency with subjective evaluation; (2) has stronger robustness and better predictive performance.
Description
technical field [0001] The invention relates to an image quality evaluation method, in particular to a full-reference image quality evaluation method based on image saliency detection, and belongs to the technical field of digital video image quality evaluation. Background technique [0002] With the continuous development of computer technology, human's demand for image processing has increased significantly, and it has been widely used in remote sensing, biomedicine, military, industrial and agricultural production, government work and other fields. Biopsychological studies have proved that for an image, humans only pay attention to the very few salient parts and ignore other areas. Image saliency detection can only focus on the salient area of an image and discard other parts, which greatly saves computing time and memory for image processing, so image saliency detection plays a very important role in image processing . [0003] Given that most digital image processin...
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 Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00G06T7/11G06T7/90
CPCG06T7/0002G06T7/11G06T7/90G06T2207/30168
Inventor 陈晨
Owner XIDIAN UNIV