We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Peningkatan Kualitas Citra Reduksi Noise Menggunakan Iterative Denoising and Backward Projection-CNN dan TFM-CLAHE pada Citra 24 Bit.
- Authors
Pardosi, Irpan Adiputra; Gohzali, Hernawati
- Abstract
Decreasing quality because of noise or abnormal contrast in images impacted objects become unclear. These problems can be caused the device is create some noise or can't produce normal contrast. The presence of noise and low dark contrast has a major impact on image quality, the large noise reduction 45% affected information in image, so quality of the reduced image becomes something that needs to be considered for large noise. In 2019 a study using Iterative Denoising and Backward Projections with CNN (IDBP-CNN) algorithm was stated able to reduce noise up to 51% with PSNR above 30 dB by ignoring the contrast of the image. Meanwhile, Triangular Fuzzy Membership–Contrast Limited Adaptive Histogram Equalization (TFM-CLAHE) algorithm is claimed able to increase image contrast with PSNR quality above 20 dB, which is better than CLAHE algorithm. Based of the tests carried out on 10 dark low-contrast images with 45% noise, combination of TFM-CLAHE algorithm and IDBP-CNN is better with an average PSNR = 31.69 dB, compared to opposite PSNR = 31.01 dB, but the average diversity of information the resulting image with combination IDBP-CNN and TFM-CLAHE is less difference than original image based on SE 3.77% compared to other is 4.75%.
- Subjects
NOISE control; ALGORITHMS; NOISE; HISTOGRAMS; IMAGE denoising
- Publication
Techno.com, 2021, Vol 20, Issue 4, p566
- ISSN
1412-2693
- Publication type
Article
- DOI
10.33633/tc.v20i4.5243