- Published: November 13, 2022
- Updated: November 13, 2022
- University / College: Queensland University of Technology (QUT)
- Language: English
- Downloads: 27
We used database is FEI FACE database, frontal images spatiallynormalized_part1 and part2, a total of 400 images, and implemented a sequenceof image processing steps to automatic normalize, equalize and crop frontalface images. The size of each image is 300×250, we put 360 images asthe training set, the remaining part as the test set. In the PSNR and SSIM (structural similarity index) numerical evaluation work, we compared the average of all testimages and evaluated their performance.
4. 2 Parametersselection We have strict requirements on thechoice of parameters, because we want to keep the same number of x and y, and under the condition, . To calculate more convenient, allthe image size to make a small change (300×250 ~300×240).
First, we test the object ispatch size P, when = 2, = 1, and select the test size of were4, 8, 16, 32 and 64 respectively (). For the SSIM andPSNR test, which tested 40 images. Fig.
5 shows the averagePSNR and SSIM for all test images when using different patch size values. In the next test, we tested the effect of the smooth parameter ? value on PSNR and SSIM, Fig. 6 showsthe average PSNR and SSIM of all test images when using different ? values (values range from 0.
3 to 2. 6). Theresults of tested PSNR and SSIM on different ?, when patch size P is 16. In the PSNR test, the best performance of PSNR can beobtained with ? of 1. 1, but when tested in SSIM, the highest value ofSSIM can be obtained when ? is 0.
9. In the following, we took ? as 1. 1 (The impact of ? onSSIM is relatively small). In the next test, we tested the effect of the number of training images on PSNR and SSIM, Fig. 7 showsthe average PSNR and SSIM of all test images when the number of training imagesis different. Theresults of tested PSNR and SSIM when the training images are taken in different numbers (100, 150, 200, 250, 300).
The result shows more training images are available when the PSNR andSSIM results can be better. The aboveexperiment tested the parameters used in our proposed SRLWR, the patch size p, ? and the number of training images. These results showed that when p is 16 and ? is 1. 1, the more training images, the better the performance. Wealso tested the performance of P and ? when the upscale factor is 4, andunder the following conditions:, where, = 4, = 1, . Fig.
8 shows the average PSNR and SSIM for all The results of tested PSNR and SSIM on different patch sizes P and ?, when the P took 32 can get bestperformance, no matter PSNR and SSIM. For the below test, when the ? took 1. 2 can get best SSIM (0. 911), but when ? took 1. 4 can get best PSNR(32. 163).
To balance the performance, we took the middle ? value 1. 3 as a following test. 4. 3 Comparison of State-of-the-Art In this part weused our proposed SRLWR to compare some state of the art methods, under the FEIface database and tested the average value of SSIM and PSNR (40 images). Ouralgorithm used the following parameters to test, p = 16, = 2 and ?= 1 when upscale factor is 2, and p = 32, = 4 and ?= 1. 3 when upscale factor is 4. For all inputimages, first use the Nearest Neighbor method downscale 4times and then upscale 2 times, which makes all the input images more blurrythan usual (in normal cases, commonly used BICUBIC downscale 2 times). ForSRLSP code, we also used the same way to deal with low-resolution trainingimages.
For VDSR code, we only took the grayscale image part. For other codeswe didn’t make any change, and numerical comparison resultis shown in Table . 1. Table 1: The result of PSNR and SSIM test on the FEI database (downscale 4 times andupscale 2 times). In the test results, we found that our proposed method can got goodscore in PSNR and SSIM numerical comparison, while the ordinarysuper-resolution algorithms does not perform well when the input image is moreblurred. Due to code problem, we can’t test SRLSP when the upscale factor is 4.
In the next experiment, we also performed the numerical comparison between PSNR and SSIM. When the input imageis just downscale 2 times with using Nearest Neighbor interpolation method, andthe comparison result is shown in Table 2. In this numerical comparison, we can see that although we proposedis not as good as the SRLSP (very close), but we can still achieve goodresults. 4. 4An intuitive image evaluative method of using normalized pseudo-color In this part, we introduced a subjective method of image evaluation, this method makes it easier to see the difference between the ground truth andthe predicted image, Fig. 9 shows effect of our evaluation method, and the specificsteps as follows, 1. Calculate the difference between the ground truth and the predictedimage (image distance). 2.
Takethe absolute value of the first step. 3. For the result of step 2, normalization from 0 to 1 is performed. 4. Using the pseudo image to represent the result of step 3.
The paper "We all test images when using different" was written by a real student and voluntarily submitted to this database. You can use this work as a sample in order to gain inspiration or start the research for your own writing. You aren't allowed to use any part of this example without properly citing it first.
If you are the author of this paper and don't want it to be used on EduPony, contact us for its removal.Ask for Removal
Cite this Essay
EduPony. (2022) 'We all test images when using different'. 13 November.
EduPony. (2022, November 13). We all test images when using different. Retrieved from https://edupony.com/we-all-test-images-when-using-different/
EduPony. 2022. "We all test images when using different." November 13, 2022. https://edupony.com/we-all-test-images-when-using-different/.
1. EduPony. "We all test images when using different." November 13, 2022. https://edupony.com/we-all-test-images-when-using-different/.
EduPony. "We all test images when using different." November 13, 2022. https://edupony.com/we-all-test-images-when-using-different/.
"We all test images when using different." EduPony, 13 Nov. 2022, edupony.com/we-all-test-images-when-using-different/.
If you have any suggestions on how to improve We all test images when using different, please do not hesitate to contact us. We want to know more: [email protected]