Erdal GÜVENOĞLU

Mobil Cihazlar Kullanılarak Elde Edilen Eğri Taranmış Görüntülerin Düzeltilmesi

Mobil cihazların yaygın olarak kullanımı ile birlikte görüntülerin yakalanması ve fotoğraflarının çekilmesi büyük ölçüde kolay hale gelmiştir. Mobil aygıtlar ile elde edilen görüntüler büyük bir oranda eğri taranmış ve düzensiz bir yapıya sahip olmaktadır. Bu eğrilikler görüntü kalitesinin düşmesine neden olmaktadır. Ayrıca elde edilecek görsel bilgilerin anlamlandırılmasını da zorlaştırmaktadır. Bu sebeplerden dolayı, mobil cihazlarda elde edilecek eğri görüntülerin düzeltilmesi önemli bir öncelik haline gelmektedir. Eğri taratılmış görüntülerin düzeltilmesi görüntü işleme tekniklerini ve matematiksel bir alt yapıyı içermektedir. Bu alanda pek çok çalışma yapılmaktadır. Bu çalışmada, eğri taranmış görüntülerin düzeltilmesi için bir yöntem sunulmuştur. Önerilen yöntem, matematiksel bir alt yapıya sahiptir. Beraberinde görüntü işleme tekniklerini içermektedir. Yöntem, mobil cihazlardan rastgele alınan perspektifi bozulmuş görüntülerin düzeltilmesini, kullanıcıya kaliteli ve iyileştirilmiş bir sonuç sunulmasını amaçlamaktadır. Elde edilen sonuç görüntüleri MSE, PSNR, SSIM ve AED gibi hata ölçüm metrikleri ile test edilmiştir. Ölçüm metriklerinden elde edilen MSE 0,0316, PSNR 23,4998, SSIM 0,9331 ve AED 0,1024 değerleri ile başarılı bir sonuca ulaşmıştır. Önerilen yöntemin literatür çalışmaları ile karşılaştırılması sağlanmış ve iyi bir başarıma sahip olduğu görülmüştür.

Anahtar Kelimeler:

Perspektif bozulma, Eğik taranmış görüntü, Görüntü işleme, Mobil tarama

Correcting Skewed Scanned Images Obtained Using Mobile Devices

With the widespread use of mobile devices, capturing and photographing images has become much easier. The images obtained with mobile devices are mostly scanned crookedly and have an irregular structure. These curvatures cause a decrease in image quality. It also makes it difficult to make sense of the visual information to be obtained. For these reasons, the correction of skewed images on mobile devices becomes an important priority. Correction of skewed scanned images involves image processing techniques and mathematical infrastructure. There are many studies in this field. In this study, a method for the correction of skew scanned images is presented. The proposed method has a mathematical background. It also includes image processing techniques. The method aims to correct perspective distorted images taken randomly from mobile devices and to provide the user with a quality and improved result. The resulting images were tested with error metrics such as MSE, PSNR, SSIM and AED. MSE 0.0316, PSNR 23.4998, SSIM 0.9331 and AED 0.1024 values obtained from the measurement metrics have achieved a successful result. The proposed method was compared with the literature studies and found to have a good performance.

Keywords:

Perspective distortion, Skew scanned image, Image processing, Mobile scanning,

PDF

___

Abdullah, S. N. H. S., Sudin, M. N., Prabuwono, A. S., ve Mantoro, T. (2012). License plate detection and segmentation using cluster run length smoothing algorithm. Journal of Information Technology Research, 5(3), 46-70.
Agrawal, N., ve Kaur, A. (2018, January). An algorithmic approach for text recognition from printed/typed text images. 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence), Noida, India.
Ahmad, R., Naz, S., ve Razzak, I. (2021). Efficient skew detection and correction in scanned document images through clustering of probabilistic hough transforms. Pattern recognition letters, 152, 93-99.
Ahmed, R., Gogate, M., Tahir, A., Dashtipour, K., Al-Tamimi, B., Hawalah, A., El-Affendi, M. A., ve Hussain, A. (2021). Novel deep convolutional neural network-based contextual recognition of Arabic handwritten scripts. Entropy, 23(3), 340.
Al-Khatatneh, A., Pitchay, S. A., ve Al-qudah, M. (2015, March). A review of skew detection techniques for document. 17th UKSim-AMSS International Conference on Modelling and Simulation (UKSim), Cambridge, UK
Alghamdi, A., Alluhaybi, D., Almehmadi, D., Alameer, K., Siddeq, S. B., ve Alsubait, T. (2021, March). Text segmentation of historical Arabic handwritten manuscripts using projection profile. 2021 National Computing Colleges Conference (NCCC), Taif, Saudi Arabia.
Ali, A. M., Benjdira, B., Koubaa, A., Boulila, W., ve El-Shafai, W. (2023). TESR: Two-Stage Approach for Enhancement and Super-Resolution of Remote Sensing Images. Remote Sensing, 15(9), 2346.
Bafjaish, S. S., Azmi, M. S., Al-Mhiqani, M. N., Radzid, A. R., ve Mahdin, H. (2018). Skew detection and correction of Mushaf Al-Quran script using hough transform. International Journal of Advanced Computer Science Applications, 9(8).
Bao, W., Yang, C., Wen, S., Zeng, M., Guo, J., Zhong, J., ve Xu, X. (2022). A novel adaptive deskewing algorithm for document images. Sensors, 22(20), 7944.
Bezmaternykh, P., ve Nikolaev, D. P. (2020, January ). A document skew detection method using fast Hough transform. Twelfth international conference on machine vision (ICMV 2019), Amsterdam, Netherlands.
Boiangiu, C.-A., Dinu, O.-A., Popescu, C., Constantin, N., ve Petrescu, C. (2020). Voting-based document image skew detection. Applied Sciences, 10(7), 2236.
Boudraa, O., Hidouci, W. K., ve Michelucci, D. (2020). Using skeleton and Hough transform variant to correct skew in historical documents. Mathematics computers in simulation, 167, 389-403.
Boukharouba, A. (2017). A new algorithm for skew correction and baseline detection based on the randomized Hough Transform. Journal of King Saud university-computer information sciences, 29(1), 29-38.
Cai, C., Meng, H., ve Qiao, R. (2021). Adaptive cropping and deskewing of scanned documents based on high accuracy estimation of skew angle and cropping value. The Visual Computer, 37, 1917-1930.
Chen, C., Seo, H., Jun, C., ve Zhao, Y. (2022). A potential crack region method to detect crack using image processing of multiple thresholding. Signal, Image Video Processing, 16(6), 1673-1681.
Chen, X., Meng, Y., Zhao, Y., Williams, R., Vallabhaneni, S. R., ve Zheng, Y. (2021, September). Learning unsupervised parameter-specific affine transformation for medical images registration. Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part IV 24, Strasbourg, France.
Chen, Y., Bahaghighat, M., Kelishomi, A. E., ve Du, J. (2023). Radon CLF: A Novel Approach for Skew Detection Using Radon Transform. Computer Systems Science Engineering, 47(1).
Chuang, C.-T., ve Lin, H.-S. (2021, October). A Effective Algorithm for Skew Correction in Text Images. International Conference on Fuzzy Theory and Its Applications (iFUZZY), Taitung, Taiwan.
Di Meo, G., Saggese, G., Strollo, A. G., ve De Caro, D. (2023). Design of Generalized Enhanced Static Segment Multiplier with Minimum Mean Square Error for Uniform and Nonuniform Input Distributions. Electronics, 12(2), 446.
Doermann, D., Liang, J., ve Li, H. (2003, August). Progress in camera-based document image analysis. Seventh International Conference on Document Analysis and Recognition, Edinburgh, UK.
Feng, H., Wang, Y., Zhou, W., Deng, J., ve Li, H. (2021). Doctr: Document image transformer for geometric unwarping and illumination correction. arXiv preprint arXiv:.12942.
Güvenoğlu, E. (2012). Optik görüntü bozulmalarının yazılımla düzeltilmesi için bir yöntem. (Doktora Tezi), Trakya Üniversitesi Fen Bilimleri Enstitüsü, Edirne.
Güvenoğlu, E. (2018). Perspektiften Kaynaklanan Bozulmaların Geometrik Olarak Düzeltilmesi İçin Bir Yöntem. Erzincan University Journal of Science Technology, 11(2), 263-276.
Güvenoğlu, E., ve Tunalı, V. (2023). ZigZag transform with Durstenfeld shuffle for fast and secure image encryption. Connection Science, 35(1), 2162000.
Hu, J., Xiawu, L., Qiao, S., Tan, W., Yin, F., Liu, T., ve Han, N. (2022). Geometric correction method for Tibetan woodcut document images. Multimedia Tools Applications, 81(11), 15609-15632.
Huang, K., Chen, Z., Yu, M., Yan, X., ve Yin, A. (2019). An efficient document skew detection method using probability model and q test. Electronics, 9(1), 55.
Jiang, B., Liu, S., Xia, S., Yu, X., Ding, M., Hou, X., ve Gao, Y. (2015, November). Video-based document image scanning using a mobile device. International Conference on Image and Vision Computing New Zealand (IVCNZ), Auckland, New Zealand.
Jundale, T. A., ve Hegadi, R. S. (2015). Skew detection and correction of Devanagari script using Hough transform. Procedia Computer Science, 45, 305-311.
Kaur, G., ve Kumar, A. (2023). Multi-level Image Enhancement for Text Recognition System using Hybrid Filters. International Journal of Intelligent Systems Applications in Engineering, 11(6s), 816–824-816–824.
Khuman, Y. L. K., Devi, H. M., ve Singh, N. A. (2021). Entropy-based skew detection and correction for printed meitei/meetei script ocr system. Materials Today: Proceedings, 37, 2666-2669.
Li, X., Liu, W., Fan, W., Sun, J., ve Satoshi, N. (2016, November). Perspective correction using camera intrinsic parameters. 13th International Conference on Signal Processing (ICSP), Chengdu, China.
Li, X., Zhang, B., Sander, P. V., ve Liao, J. (2019, June). Blind geometric distortion correction on images through deep learning. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.
Lu, X. X. (2018). A review of solutions for perspective-n-point problem in camera pose estimation. Journal of Physics: Conference Series, 1087(5), 052009.
Luqman, M. M., Gomez-Krämer, P., ve Ogier, J.-M. (2014, August). Mobile phone camera-based video scanning of paper documents. Camera-Based Document Analysis and Recognition: 5th International Workshop, Washington, DC, USA.
Mohammad, K., Qaroush, A., Washha, M., Agaian, S., ve Tumar, I. (2021). An adaptive text-line extraction algorithm for printed Arabic documents with diacritics. Multimedia Tools Applications, 80, 2177-2204.
Mukhopadhyay, P., ve Chaudhuri, B. B. (2015). A survey of Hough Transform. Pattern Recognition, 48(3), 993-1010.
Ouwayed, N., Belaid, A., ve Auger, F. (2009). Skew angle estimation of scanned handwritten Arabic documents using a time-frequency analysis of the projection histograms. Traitement DU Signal, 26(4), 307-319.
Peake, G., ve Tan, T. (1997, October). A general algorithm for document skew angle estimation. Proceedings of International Conference on Image Processing, Santa Barbara, CA, USA.
Postl, W. (1986, October). Detection of linear oblique structures and skew scan in digitized documents. Proc. Int. Conf. on Pattern Recognition, Paris, France.
Romanengo, C., Biasotti, S., ve Falcidieno, B. (2022). Hough transform for detecting space curves in digital 3d models. Journal of Mathematical Imaging and Vision, 64(3), 284-297.
Salagar, R., ve Patil, P. B. (2020, March). Application of RLSA for skew detection and correction in Kannada text images. Fourth International Conference on Computing Methodologies and Communication (ICCMC), Erode, India.
Singh, P., ve Sharma, D. V. (2023). Pre-Processing of Mobile Camera Captured Images for OCR. International Journal of Intelligent Systems Applications in Engineering, 11(2s), 147-155.
Sonkusare, M., Gupta, R., ve Moghe, A. (2021). A Review on Character Segmentation Approach for Devanagari Script. Intelligent Systems: Proceedings of SCIS, 181-189.
Teplyakov, L., Kaymakov, K., Shvets, E., ve Nikolaev, D. (2021, January). Line detection via a lightweight CNN with a Hough layer. Thirteenth International Conference on Machine Vision, Rome, Italy.
Tinungki, G. M., ve Nurwahyu, B. (2020). The implementation of Google Classroom as the e-learning platform for teaching Non-Parametric Statistics during COVID-19 pandemic in Indonesia. International Journal of Advanced Science Technology, 29(4), 5793-5803.
Trstenjak, B., Mikac, S., ve Trstenjak, J. (2018). The Framework for Fast Skew Angle Detectıon and Auto Correctıon Of Scanned Documents. Annals of DAAAM Proceedings, 29.
Wu, L., Shang, Q., Sun, Y., ve Bai, X. (2019). A self-adaptive correction method for perspective distortions of image. Frontiers of Computer Science, 13(3), 588-598.
Zheng, W., Yu, H., ve Lu, Z. (2021). Two-step affine transformation prediction for visual object tracking. IEEE Access, 9, 36512-36521.
Zohrevand, A., Sadri, J., Imani, Z., ve Yeganezad, M. R. (2019, March). Line segmentation in Persian handwritten documents based on a novel projection histogram method. 4th International Conference on Pattern Recognition and Image Analysis (IPRIA), Tehran, Iran.