A mobile phone anti-shake function evaluation method based on image blur

Abstract

The invention discloses a mobile phone anti-shake function evaluation method based on image ambiguity. By fixing the position of the mobile phone, taking a static image, and then turning on the vibrating table, a group of images are respectively taken when the mobile phone anti-shake function is turned on and off, and then , calculate image blur based on deep learning method, use deep learning deblurring network for feature extraction, calculate Wasserstein distance of feature vector, get blur amount of two groups of images, and finally generate image blur evaluation result according to blur amount. In the method, the entire image will participate in the calculation of the blur degree. Compared with the existing evaluation method that uses partial images for calculation, it avoids the influence of different regions on the calculation results, and the calculation is more reasonable and the result is more accurate. At the same time, the calculated The ambiguity obtained makes a detailed and clear division of the anti-shake function of mobile phones, which improves the objectivity of the evaluation results and makes it easier for users to understand and grasp.

Description

technical field [0001] The invention relates to an evaluation method, in particular to an evaluation method based on image blur for the anti-shake function of a mobile phone, and belongs to the field of evaluation of mobile phones. Background technique [0002] The camera function is an important part of the mobile phone, and its shooting quality directly affects the sales of smart phones, so manufacturers are constantly updating and improving the shooting quality. The anti-shake function of mobile phones is a necessary prerequisite for taking high-quality photos, especially for smart phone products, its effect is more prominent and obvious. [0003] The current evaluation process for the anti-shake function of mobile phones mainly includes: [0004] 1) Place the mobile phone on the vibrating table, change the Shutter Speed ​​(SS) of the mobile phone according to the shooting requirements stated in the anti-shake test standard "CIPA DC-X011-Translation-2014", and take multi...

AI Technical Summary

Problems solved by technology

[0015] 1) The content of the standard focuses on the description of the evaluation operation process, and the definition of the calculation part of the evaluation result is not clear enough, and there is ambiguity, which leads to different evaluation results obtained by users using different calculation methods in the actual use process;

[0016] 2) The evaluation method described in the standard is only for mobile phones that use optical image stabilization, and the evaluation object is the amount of blur in the captured picture, which reduces the universality of the evaluation method. Some smart phones that use electronic image stabilization, digital image stabilization, etc. cannot be evaluated effectively;

[0017] 3) The standard evaluation results are anti-shake levels, such as level 1, level 3, etc., which are too general, and non-professionals cannot correctly interpret the results, which will affect users' judgment on mobile phone anti-shake performance

Images