identification of weed in agricultural field by using video matting and image segmentation
R.Raj Bharath,Heera
Published in International Journal of Advanced Research in Computer Science Engineering and Information Technology
ISSN: 2321-3337 Impact Factor:1.521 Volume:4 Issue:3 Year: 29 April,2016 Pages:991-998
Abstract
The main objective of the paper is to evaluate a software solution for automatic identification of weed from original crops .The identification process is achieved by implementing image segmentation and Video Matting techniques. The aim of the project is to detect and classify weeds from different species of huge cultivated crops in our country like paddy, wheat, brinjal, tomato other cash crops and oil seeds etc. The plant identification species were taken for our approach is first taken as trained set of data by collecting its details such as its size, color, texture and shape. The trained data is processed till its complete parameters are identified clearly in order to classify it from other crops. The experimental results indicate the proposed approach can recognize and classify the crop identification with a little computational effort. The proposed method mainly focuses on analyzing and identification of weeds which is attained by identifying the non similar parameters to the trained set of data. The inputs for the system are feed as videos of the cultivated crops which is processed by video matting technique and converted into frames by frame diffusion algorithm .The related structure and segmenting parameters like size, color, texture and shape were collected as testing set of data from the converted frames . The segmented parameters where analyzed with the available training set of data and the result is generated as to identify the weed from the original crops.
Kewords
Image Segmentation, Video Matting, Support Vector Machine, One Class Support Vector Machine, Foreground Segmentation.
Reference
[1] J. L. Bentley, “Integrated Foreground Segmentation and Boundary Matting for live Videos,” Commun. ACM, vol. 18, no. 9, pp. 509–517, 1975. [2] R. F. Sproull, “Geodesic matting:A framewrok for last interactive image and video segmentation and matting.,” Algorithmic a, vol. 6, no. 4, pp. 579–589, 1991. [3] S.Korman and S.Avidan ,” Realtime background subtraction from dynamic scenes”, in proc.ICCV 2011,pp.1607-1614. [4] C.Barnes , E.Shectman, A.Finkelstein , and D.B .Goldman, “Patch match: A randomized correspondence algorithm for structural image editing”, ACM Trans. Graph.,vol.28, no. 3,pp.1-8,2009. [5] S.A.Ramakanth and R.V.Babu, “Defocus video matting.,”Comput.Med.Imag.Graph” vol.38,no.1,pp.49-56,2014. [6] Y.Hacohen, E.Shechtman, D.B.Goldman, and D.Lischinski,”Non-rigid dense correspondence with applications for image enhancement”, ACM Trans. Graph.,vol.30, no.4,pp.1-70,2011. [7] A. Yilmaz, O. Javed, and M. Shah, “background subtraction for freely moving cameras,” ACM Comput. Surv., vol. 38, Dec. 2012 [8] S. Hinz, R. Bamler, and U. Stilla, “Scribble tracker: A matting based approach for robust tracking,” ISPRS J. Photogramm. Remote Sens., vol. 61, no. 3–4, pp. 135–280, 2011 [9] I. Szottka and M. Butenuth, “Automatic real time video matting using time-of-flight camera and multichannel poisson equations,” in Proc. 2011 Joint Urban Remote Sensing Event (JURSE), Apr. 2011, pp. 13–16. [10] K. Palaniappan, F. Bunyak, P. Kumar, I. Ersoy, S. Jaeger, K. Ganguli, A. Haridas, J. Fraser, R. Rao, and G. Seetharaman, “Video sanpcut:robust video object cutout using localized classifiers,” in Proc. 13th Conf. Information Fusion (FUSION),Jul. 2010, pp.
