BPG-based compression analysis of Poisson-noisy medical images

Victoriia Naumenko, Bogdan Kovalenko, Volodymyr Lukin

Abstract


The subject matter is lossy compression using the BPG encoder for medical images with varying levels of visual complexity, which are corrupted by Poisson noise. The goal of this study is to determine the optimal parameters for image compression and select the most suitable metric for identifying the optimal operational point. The tasks addressed include: selecting test images sized 512x512 in grayscale with varying degrees of visual complexity, encompassing visually intricate images rich in edges and textures, moderately complex images with edges and textures adjacent to homogeneous regions, and visually simple images primarily composed of homogeneous regions; establishing image quality evaluation metrics and assessing their performance across different encoder compression parameters; choosing one or multiple metrics that distinctly identify the position of the optimal operational point; and providing recommendations based on the attained results regarding the compression of medical images corrupted by Poisson noise using a BPG encoder, with the aim of maximizing the restored image’s quality resemblance to the original. The employed methods encompass image quality assessment techniques employing MSE, PSNR, MSSIM, and PSNR-HVS-M metrics, as well as software modeling in Python without using the built-in Poisson noise generator. The ensuing results indicate that optimal operational points (OOP) can be discerned for all these metrics when the compressed image quality surpasses that of the corresponding original image, accompanied by a sufficiently high compression ratio. Moreover, striking a suitable balance between the compression ratio and image quality leads to partial noise reduction without introducing notable distortions in the compressed image. This study underscores the significance of employing appropriate metrics for evaluating the quality of compressed medical images and provides insights into determining the compression parameter Q to attain the BPG encoder’s optimal operational point for specific images. Conclusions. The scientific novelty of the findings encompasses the following: 1) the capability of all metrics to determine the OOP for images of moderate visual complexity or those dominated by homogeneous areas; MSE and PSNR metrics demonstrating superior results for images rich in textures and edges; 2) the research highlights the dependency of Q in the OOP on the average image intensity, which can be reasonably established for a given image earmarked for compression based on our outcomes. The compression ratios for images compressed at the OOP are sufficiently high, further substantiating the rationale for compressing images in close proximity to the OOP.

Keywords


lossy image compression; BPG; Poisson noise; optimal operation point

Full Text:

PDF

References


Ng, K. H., Faust, O., Sudarshan, V., & Chattopadhyay, S. Data Overloading in Medical Imaging: Emerging Issues, Challenges and Opportunities in Efficient Data Management. Journal of Medical Imaging and Health Informatics, 2015, vol. 5, iss. 4, pp. 755–764. DOI: 10.1166/jmihi.2015.1449.

Patidar, G., Kumar, S., & Kumar, D. A Review on Medical Image Data Compression Techniques. 2nd International Conference on Data, Engineering and Applications (IDEA), Bhopal, India, 2020, pp. 1-6. DOI: 10.1109/IDEA49133.2020.9170679.

Kumar, S., & Kumar, D. Comparative Analysis and Performance Evaluation of Medical Image Compression Method for Telemedicine. 2nd International Conference on Data, Engineering and Applications (IDEA), Bhopal, India, 2020, pp. 1-5. DOI: 10.1109/IDEA49133.2020.9170724.

Hussain, A. J., Al-Fayadh, A., & Radi, N. Image Compression Techniques: A Survey in Lossless and Lossy algorithms. Neurocomputing, 2018, vol. 300, pp. 44-49. DOI: 10.1016/j.neucom.2018.02.094.

Kryvenko, S., Lukin, V., Krylova, O., Kryvenko, L., & Egiazarian, K. A Fast Method of Visually Lossless Compression of Dental Images. Applied Sciences, 2021, vol. 11, iss. 1, article no. 135. DOI: 10.3390/app11010135.

Kumar, S. N., Haridhas, A. K., Fred, A. L., & Varghese, P. S. Analysis of Lossy and Lossless Compression Algorithms for Computed Tomography Medical Images Based on Bat and Simulated Annealing Optimization Techniques. Computational Intelligence Methods for Super-Resolution in Image Processing Applications. Springer, Cham., 2021, pp. 99-133. DOI: 10.1007/978-3-030-67921-7_6.

Moura, L., Furuie, S. S., Gutierrez, M. A., Tachinardi, U., Rebelo, M. S., Alcocer, P., & Melo, C. P. Lossy compression techniques, medical images, and the clinician. MD Computing : computers in medical practice, 1996, vol. 13, iss. 2, pp. 155-172. PMID: 8684278.

Sultana, Z., Nahar, L., Tasnim, F., Hossain, M. S., & Andersson, K. Lossy Compression Effect on Color and Texture Based Image Retrieval Performance. Intelligent Computing & Optimization. ICO 2022. Lecture Notes in Networks and Systems, Springer, Cham., 2023, vol. 569, pp. 1159-1167. DOI: 10.1007/978-3-031-19958-5_108.

European Society of Radiology (ESR). Usability of irreversible image compression in radiological imaging. A position paper by the European Society of Radiology (ESR). Insights into Imaging, 2011, vol. 2, iss. 2, pp. 103-115. DOI: 10.1007/s13244-011-0071-x.

Barannik, V., Sidchenko, S., Barannik, D., Yermachenkov, A., Savchuk, M., & Pris, G. Video images compression method based on floating positional coding with an unequal codograms length. Radioelectronic and Computer Systems, 2023, no. 1, pp. 134-146. DOI: 10.32620/reks.2023.1.11.

Liu, F., Hernandez-Cabronero, M., Sanchez, V., Marcellin, M. W., & Bilgin, A. The Current Role of Image Compression Standards in Medical Imaging. Information, 2017, vol. 8, iss. 4, article no. 131. DOI: 10.3390/info8040131.

Guidance for the content and review of 510(K) Notifications for Picture Archiving and Communications Systems (PACS) and Related Devices. Food and Drug Administration. 2023. Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K211257. (accessed 12.06.2023).

Lukin, V., Bataeva, E., & Abramov, S. Saliency map in image visual quality assessment and processing. Radioelectronic and computer systems, 2023, no. 1, pp. 112-121. DOI: 10.32620/reks.2023.1.09.

Bondžulić, B., Pavlović, B., Stojanović, N., & Petrović, V. Picture-wise just noticeable difference prediction model for JPEG image quality assessment. Vojnotehnicki glasnik, 2022, vol. 70, iss. 1, pp. 62-86. DOI: 10.5937/vojtehg70-34739.

Gooden, D. S. Legal aspects of image compression. In Proceedings of the American Assoc. of Physicists in Medicine (AAPM). 35th Annual Meeting, Washington, DC, 1993, pp. 8-12.

Koff, D., Bak, P., Brownrigg, P. et al. Pan-Canadian evaluation of irreversible compression ratios (“lossy” compression) for development of national guidelines. Journal of digital imaging, 2009, vol. 22, pp. 569-578. DOI: 10.1007/s10278-008-9139-7.

Prasath, V. B. S. Quantum Noise Removal in XRay Images with Adaptive Total Variation Regularization, Informatica, 2017, vol. 28, iss. 3, pp. 505-515. DOI: 10.15388/Informatica.2017.141.

Le, T., Chartrand, R., & Asaki, T. J. A Variational Approach to Reconstructing Images Corrupted by Poisson Noise. Journal of Mathematical Imaging and Vision, 2007, vol. 3, pp. 257-263. DOI: 10.1007/s10851-007-0652-y.

Li, B., Yang, R., & Jiang, H. Remote-Sensing Image Compression Using Two-Dimensional Oriented Wavelet Transform. IEEE Transactions on Geoscience and Remote Sensing, 2011, vol. 49, iss. 1, pp. 236-250. DOI: 10.1109/TGRS.2010.2056691.

Al-Chaykh, O. K., & Mersereau, R. M. Lossy compression of noisy images. IEEE Transactions on Image Processing, 1998, vol. 7, iss. 12, pp. 1641-1652. DOI: 10.1109/83.730376.

Shahnaz, R., Walkup, J. F., & Krile T. F. Image Compression in Signal-Dependent Noise. Applied Optics, 1999, vol. 38, iss. 26, pp. 5560-5567. DOI: 10.1364/AO.38.005560.

Chang, S. G., Yu, B., & Vetterli, M. Adaptive wavelet thresholding for image denoising and compression. IEEE Trans. on Image Processing, 2000, vol. 9, iss. 9, pp. 1532-1546. DOI: 10.1109/83.862633.

Lim, S. H. Characterization of Noise in Digital Photographs for Image Processing. Proceedings of Digital Photography II, 2006, vol. 60690O, pp. 219-228. DOI: 10.1117/12.655915.

Lukin, V., Kovalenko, B., Kryvenko, S., Naumenko, V., & Vozel, B. Prediction of Optimal Operation Point Existence and Its Parameters in BPG-Based Automatic Lossy Compression of Noisy Images. Current Overview on Science and Technology Research, 2022, vol. 9, pp. 1-36. DOI: 10.9734/bpi/costr/v9/4316A.

Kovalenko, B., Lukin, V., Kryvenko, S., Naumenko, V., & Vozel, B. BPG-Based Automatic Lossy Compression of Noisy Images with the Prediction of an Optimal Operation Existence and Its Parameters. Applied Sciences, 2022, vol. 12, iss. 15, article no. 7555. DOI: 10.3390/app12157555.

Naumenko, V., Lukin, V., & Krivenko, S. Analysis of Noisy Image Lossy Compression by BPG. Integrated Computer Technologies in Mechanical Engineering – 2021. ICTM 2021. Lecture Notes in Networks and Systems, Springer, Cham, 2022, vol. 367, pp. 911-923. DOI: 10.1007/978-3-030-94259-5_71.

Makarichev, V., Lukin, V., & Brysina, I. Progressive DCT-based coder and its comparison to atomic function based image lossy compression. IEEE 16th International Conference on Advanced Trends in Radioelectronics, Telecommunications and Computer Engineering (TCSET), Lviv-Slavske, Ukraine, 2022, pp. 01-06, DOI: 10.1109/TCSET55632.2022.9766871.

Li, F., Krivenko, S., & Lukin, V. A Two-step Procedure for Image Lossy Compression by ADCTC with a Desired Quality. IEEE 11th International Conference on Dependable Systems, Services and Technologies (DESSERT), Kyiv, Ukraine, 2020, pp. 307-312. DOI: 10.1109/DESSERT50317.2020.9125000.

Wang, Z., Simoncelli, E. P., & Bovik, A. C. Multiscale structural similarity for image quality assessment. The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003, vol. 2, Pacific Grove, CA, USA, 2003, pp. 1398-1402. DOI: 10.1109/ACSSC.2003.1292216.

Lukin, V. V., Krivenko, S. S., Zriakhov, M. S., Ponomarenko, N. N., Abramov, S. K., Kaarna, A., & Egiazarian, K. Lossy compression of images corrupted by mixed Poisson and additive Gaussian noise. 2009 International Workshop on Local and Non-Local Approximation in Image Processing, Tuusula, Finland, 2009, pp. 33-40. DOI: 10.1109/LNLA.2009.5278407.

Azzari, L., Borges, L. R., & Foi, A. Modeling and Estimation of Signal-Dependent and Correlated Noise. Denoising of Photographic Images and Video. Advances in Computer Vision and Pattern Recognition. Springer, Cham, 2018, pp. 1-36. DOI: 10.1007/978-3-319-96029-6_1.




DOI: https://doi.org/10.32620/reks.2023.3.08

Refbacks

  • There are currently no refbacks.