An Approach to Counting Legumes Using Coordinate Features in Image Processing

Abstract

: Object counting is a process used for different tasks in food, medical, industry and daily life. In this study, image processing-based systems have been examined and applications have been made to perform the object counting process. For the applications, an experimental setup has been designed in which the objects can flow over a silo and proceed in an inclined manner over the spreader. An industrial camera and lens system that can observe the flow of objects were used. Only backlighting was used to monitor the flow of objects. The flow rate of objects and the angle of inclination through which objects can flow out of the spreader can be changed. In the industrial camera, fps, pixel frequency, exposure time, image resolution, area of interest selection, color and mono image acquisition, how many bits the pixels in the image will be sampled can be selected by the user. Python software and OpenCV library were used for algorithm design. For the object counting, 100 chickpea seeds were determined and the video was recorded from the camera for a certain period of time. By applying a background subtraction process on the video, a second video was obtained in which only the objects can be seen as white. The smallest rectangular coordinate values bounding the object and the object centroid coordinate values were obtained in the video converted into binary format. Methods have been developed for counting by drawing single and double virtual lines in the video. The advantages and disadvantages of using coordinate values and virtual lines for counting have been discussed as a result of this study.

Keywords

• Object Counting
• Image Processing
• Camera
• Python
• OpenCV

GÖRÜNTÜ İŞLEMEDE NESNE KOORDİNAT ÖZELLİKLERİNİ KULLANARAK BAKLİYAT SAYMA İŞLEMİNE BİR YAKLAŞIM

Abstract

Nesne sayma, gıda, medikal, endüstri ve günlük yaşamda farklı görevler için kullanılan bir süreçtir. Bu çalışmada, nesne sayma işleminin gerçekleştirilebilmesi için görüntü işleme tabanlı sistemler incelenmiş ve uygulamalar yapılmıştır. Uygulamalar için nesnelerin bir silo üzerinden akarak serici üzerinden eğimli bir şekilde ilerleyebileceği bir deney düzeneği tasarlanmıştır. Nesnelerin akışını gözlemleyebilecek endüstriyel bir kamera ve lens sistemi kullanılmıştır. Nesnelerin akışını izleyebilmek için sadece arka aydınlatma kullanılmıştır. Nesnelerin akış hızı ve nesnelerin serici çıkışı akabileceği eğim açısı değiştirilebilmektedir. Endüstriyel kamerada fps, piksel frekansı, pozlama süresi, görüntünün çözünürlüğü, ilgilenen alan seçimi, renkli ve renksiz görüntü alımı, görüntüdeki piksellerin kaç bit ile örnekleneceği kullanıcı tarafından seçilebilmektedir. Algoritma tasarımı için Python yazılımı ve OpenCV kütüphanesi kullanılmıştır. Nesne sayımı için 100 adet nohut tanesi belirlenmiş ve belirli bir süre kameradan video kaydı elde edilmiştir. Video üzerinde arka plan çıkarma işlemi uygulanarak sadece nesnelerin beyaz renk olarak görülebileceği ikinci bir video elde edilmiştir. Binary formata dönüştürülen videoda nesne çevresi sınırlayıcı en küçük dikdörtgen koordinat değerleri ve nesne ağırlık merkezi koordinat değerleri elde edilmiştir. Video görüntüsü üzerinde sabit tek ve çift sanal çizgiler çekilerek sayma işleminin gerçekleştirileceği metotlar geliştirilmiştir. Koordinat değerlerinin ve sanal çizgilerin sayma işlemi için kullanılmasında ortaya çıkan avantaj ve dezavantajlar bu çalışma sonucunda tartışılmıştır.

Keywords

• Nesne Sayma
• Görüntü İşleme
• Kamera
• Python
• OpenCV

References

1. Aich, S., Stavness, I., "Leaf counting with deep convolutional and deconvolutional networks", Proceedings of the IEEE International Conference on Computer Vision Workshops, Venice, Italy, 2080-2089, 22-29 Oct. 2017.
2. Aich, S., Josuttes, A., Ovsyannikov, I., Strueby, K., Ahmed, I., Duddu, H. S., Pozniak, C., Shirtliffe, S., Stavness, I., "Deepwheat: Estimating phenotypic traits from crop images with deep learning", IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Tahoe, NV, USA, 323-332, 12-15 March 2018.
3. Antonini, G., Thiran, J.-P. J. I. T. o. C., Technology, S. f. V., 2006, "Counting pedestrians in video sequences using trajectory clustering", IEEE Transactions on Circuits and Systems for Video Technology, Vol. 16, No. 8, pp. 1008-1020.
4. Arteta, C., Lempitsky, V., Noble, J. A., Zisserman, A., "Learning to detect cells using non-overlapping extremal regions", International Conference on Medical Image Computing and Computer-Assisted Intervention, Berlin, Heidelberg, 348-356, 2012.
5. Barbedo, J. G. A., "Counting clustered soybean seeds", 2012 12th International Conference on Computational Science and Its Applications, Salvador, Brazil, 142-145, 2012.
6. Baygin, M., Karakose, M., Sarimaden, A., Akin, E. J. a. p. a., 2018, "An image processing based object counting approach for machine vision application", arXiv preprint arXiv:1802.05911.
7. Chan, A. B., Liang, Z.-S. J., Vasconcelos, N., "Privacy preserving crowd monitoring: Counting people without people models or tracking", IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, AK, USA, 1-7, 2008.
8. Chan, A. B., Vasconcelos, N. J. I. T. o. I. P., 2011, "Counting people with low-level features and Bayesian regression", IEEE Transactions on Image Processing , Vol. 21, No. 4, pp. 2160-2177.

9. Chen, K., Gong, S., Xiang, T., Change Loy, C., "Cumulative attribute space for age and crowd density estimation", Proceedings of the IEEE conference on computer vision and pattern recognition, Portland, OR, USA, 2467-2474, 2013.
10. Lin, Z., Davis, L. S. J. I. T. o. P. A., Intelligence, M., 2010, "Shape-based human detection and segmentation via hierarchical part-template matching", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 32, No. 4, pp. 604-618.
11. Mazurek, P., YOLO-Object-Counting-API, https://github.com/tugot17/YOLO-Object-Counting-API, Ziyaret Tarihi: 28.09.2020
12. Onoro-Rubio, D., López-Sastre, R. J.,"Towards perspective-free object counting with deep learning", European Conference on Computer Vision, Amsterdam, Netherlands, 615-629, 11-14 October, 2016.
13. OpenCV, 2020a, Image Thresholding, https://docs.opencv.org/3.4/d7/d4d/tutorial_py_thresholding.html, Ziyaret Tarihi: 28.09.2020.
14. OpenCV, 2020b, Motion Analysis, https://docs.opencv.org/master/de/de1/group__video__motion.html, Ziyaret Tarihi: 28.09.2020.
15. OpenCV, 2020c, Contours : Getting Started, https://docs.opencv.org/master/d4/d73/tutorial_py_ contours_begin.html, Ziyaret Tarihi: 28.09.2020.
16. OpenCV, 2020d, Structural Analysis and Shape Descriptors, https://docs.opencv.org/3.4/d3/dc0/group__ imgproc__shape.html#ga103fcbda2f540f3ef1c042d6a9b35ac7, Ziyaret Tarihi: 28.09.2020.
17. Özlü, A., 2020, Tensorflow Object Counting API, https://github.com/ahmetozlu/tensorflow_object_ counting_api, Ziyaret Tarihi: 28.09.2020.
18. Öztürk, Ş., Özkaya, U., 2020, "Skin Lesion Segmentation with Improved Convolutional Neural Network", Journal of Digital Imaging.
19. Pandit, A., Rangole, J., Shastri, R., Deosarkar, S., "Vision system for automatic counting of silkworm eggs", International Conference on Information Communication and Embedded Systems (ICICES2014), 1-5, Chennai, India, 2014.
20. Perera, P., Fernando, W., Herath, H., Godaliyadda, G., Ekanayake, M., Wijayakulasooriya, J.,"A generic object counting algorithm under partial occlusion conditions", IEEE 8th International Conference on Industrial and Information Systems, Peradeniya, Sri Lanka, 554-559, 17-20 Dec. 2013.
21. Rabaud, V., Belongie, S., "Counting crowded moving objects", IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06), New York, NY, USA , 705-711, 17-22 June 2006.
22. Raman, M. S., Sukanya, M., 2012, "A novel labelling algorithm for object counting", 2012 Third International Conference on Computing, Communication and Networking Technologies (ICCCNT'12), Coimbatore, India , 1-7, 26-28 July 2012.
23. Ryan, D., Denman, S., Fookes, C., Sridharan, S., "Crowd counting using multiple local features", Digital Image Computing: Techniques and Applications, Melbourne, VIC, Australia, 81-88, 1-3 Dec. 2009.
24. Thammasorn, P., Boonchu, S., Kawewong, A., "Real-time method for counting unseen stacked objects in mobile", IEEE International Conference on Image Processing, Melbourne, VIC, Australia, 4103-4107, 15-18 Sept. 2013.
25. Wang, M., Wang, X., "Automatic adaptation of a generic pedestrian detector to a specific traffic scene", CVPR 2011, Providence, RI, USA , 3401-3408, 20-25 June 2011.
26. Zhang, C., Li, H., Wang, X., Yang, X., "Cross-scene crowd counting via deep convolutional neural networks", Proceedings of the IEEE conference on computer vision and pattern recognition, Boston, MA, USA , 833-841, 7-12 June 2015.
27. Zhang, Y., Zhou, D., Chen, S., Gao, S., Ma, Y., "Single-image crowd counting via multi-column convolutional neural network", Proceedings of the IEEE conference on computer vision and pattern recognition, Las Vegas, NV, USA, 589-597, 27-30 June 2016.