평활화 알고리즘에 따른 자궁경부 분류 모델의 성능 비교 연구
A Performance Comparison of Histogram Equalization Algorithms for Cervical Cancer Classification Model
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
We developed a model to classify the absence of cervical cancer using deep learning from the cervical image to which the histogram equalization algorithm was applied, and to compare the performance of each model. A total of 4259 images were used for this study, of which 1852 images were normal and 2407 were abnormal. And this paper applied Image Sharpening(IS), Histogram Equalization(HE), and Contrast Limited Adaptive Histogram Equaliza�tion(CLAHE) to the original image. Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity index for Measuring image quality(SSIM) were used to assess the quality of images objectively. As a result of assessment, IS showed 81.75dB of PSNR and 0.96 of SSIM, showing the best image quality. CLAHE and HE showed the PSNR of 62.67dB and 62.60dB respectively, while SSIM of CLAHE was shown as 0.86, which is closer to 1 than HE of 0.75. Using ResNet-50 model with transfer learning, digital image-processed images are classified into normal and abnormal each. In conclusion, the classification accuracy of each model is as follows. 90.77% for IS, which shows the highest, 90.26% for CLAHE and 87.60% for HE. As this study shows, applying proper digital image proces
Abstract
We developed a model to classify the absence of cervical cancer using deep learning from the cervical image to which the histogram equalization algorithm was applied, and to compare the performance of each model. A total of 4259 images were used for this study, of which 1852 images were normal and 2407 were abnormal. And this paper applied Image Sharpening(IS), Histogram Equalization(HE), and Contrast Limited Adaptive Histogram Equaliza�tion(CLAHE) to the original image. Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity index for Measuring image quality(SSIM) were used to assess the quality of images objectively. As a result of assessment, IS showed 81.75dB of PSNR and 0.96 of SSIM, showing the best image quality. CLAHE and HE showed the PSNR of 62.67dB and 62.60dB respectively, while SSIM of CLAHE was shown as 0.86, which is closer to 1 than HE of 0.75. Using ResNet-50 model with transfer learning, digital image-processed images are classified into normal and abnormal each. In conclusion, the classification accuracy of each model is as follows. 90.77% for IS, which shows the highest, 90.26% for CLAHE and 87.60% for HE. As this study shows, applying proper digital image proces
참고문헌 (33)
- Annual report of cancer statistics in Korea in 2017, "Korea Central Cancer Registry, National Cancer Center", Ministry of Health and Welfare, 33~ (2019)
- Coppleson, LW, "Estimation of the screening error rate from the observed detection rates in repeated cervical cytology", American Journal of Obstetrics and Gynecology, 119(7), pp.953~958 (1974)
- Zarchi MK, "Value of colposcopy in the early diagnosis of cervical cancer in patients with abnormal pap smears at Shahid Sadoughi Hospital, Yazd", Asian Pacific Journal of Cancer Prevention, 12(12), pp.3439~3441 (2011)
- Alyafeai Z, "A fully-automated deep learning pipeline for cervical cancer classification", Expert Systems with Applications, 141 (2020)
- Wu M, "Automatic classification of cervical cancer from cytological images by using convolutional neural network", Bioscience Reports, 38(6), pp.1~9 (2018)
- Kudva V, "Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening", Lecture Notes in Electrical Engineering, 614, pp.299~312 (2020)
- Lange H, "Automatic detection of multi-level acetowhite regions in RGB color images of the uterine cervix", Medical Imaging 2005: Image Processing, 5747, pp.1004~ (2005)
- Lee Y, "Deep Learning-based Feature Extraction for Medical Image Analysis", pp.1~12 (2014)
- Jeelani H, "Image quality affects deep learning reconstruction of MRI", International Symposium on Biomedical Imaging, pp.357~360 (2018)
- Guo P, "Deep learning for assessing image focus for automated cervical cancer screening", 2019 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2019, pp.2019~2022 (2019)
- Ulku EE, "Computer aided brain tumor detection with histogram equalization and morphological image processing techniques", ICECCO 2013, pp.48~51 (2013)
- Ullah Z, "A hybrid image enhancement based brain MRI images classification technique", Medical Hypotheses, 143, pp.109922~ (2020)
- Ziaei A, "A novel approach for contrast enhancement in biomedical images based on histogram equalization", BMEI 2008, 1, pp.855~858 (2008)
- Suralkar SR, "Enhancement of Images Using Contrast Image Enhancement Techniques", International Journal of Recent Advances in Engineering & Technology, 8(3), pp.16~20 (2020)
- Singh RP, "Histogram Equalization: A Strong Technique for Image Enhancement", International Journal of Signal Processing, Image Processing and Pattern Recognition, 8(8), pp.345~352 (2015)
- Gonzalez RC, "Digital Image Processing", pp.394~460 (2008), Pearson Education, Inc
- Russo F, "Piecewise Linear Model-Based Image Enhancement", EURASIP J. Adv. Signal Process, pp.983173~ (2004)
- Yun JW, "Deep Residual Learning for Image Recognition", Enzyme and Microbial Technology, 19(2), pp.107~117 (1996)
- Kurt B, "Medical images enhancement by using anisotropic filter and CLAHE", INISTA 2012 -International Symposium on INnovations in Intelligent Sys-Tems and Applications, pp.1~4 (2012)
- 김민기, "ResNet 모델을 이용한 눈 주변 영역의 특징 추출 및 개인 인증", 멀티미디어학회논문지, 22(12), pp.1347~1355 (2019), 한국멀티미디어학회[DOI]
- Weiss K, "A survey of transfer learning", Journal of Big Data. Springer International Publishing, 3(1) (2016)
- Horé A, "Image quality metrics: PSNR vs. SSIM", International Conference on Pattern Recognition, pp.2366~2369 (2010)
- 강광수, "MRI 영상의 PSNR 평가", 한국방사선학회논문지, 3(4), pp.13~20 (2009), 한국방사선학회
- Wang Z, "Image quality assessment: From error visibility to structural similarity", IEEE Transactions on Image Processing, 13(4), pp.600~612 (2004)
- Nevriyanto A, "Enhancement, and Standard Median Filter (Noise Removal) with Pixel-Based and Human Visual System-Based Measurements", International Conference on Electrical Engineering and Computer Science, 1(1), pp.114~119 (2017)
- Kugelman J, "Effect of Altered OCT Image Quality on Deep Learning Boundary Segmentation", IEEE Access, 8, pp.43537~43553 (2020)
- Kaur H, "MRI brain image enhancement using Histogram Equalization techniques", Proceedings of the 2016 IEEE International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2016, 1, pp.770~773 (2016)
- Srinivasan Y, "A probabilistic approach to segmentation and classification of neoplasia in uterine cervix images using color and geometric features", Progress in Biomedical Optics and Imaging - Proceedings of SPIE, 5747(2), pp.995~1003 (2005)