Identification of foliar soybean diseases using local descriptors PIRES, R. D. L. ; KANASHIRO, W. E. S. ; GONÇALVES, W. N. ; MACHADO, B. B. ; ARRUDA, M. S. ; ORUE, J. P. M. XI Workshop de Visão Computacional, p. 242-247, 2015
The identification of foliar diseases is very important in the grain production. In the last decade, a great number of soybean leaf diseases have impacted the croping in Brazil. This paper presents a local feature approach to identify soybean leaf diseases. We compared different local descriptors, such as SIFT, Dense SIFT and SURF over a Bag-of-Visual-Words model. Experimental results demonstrated that local descriptors are efficient in performing the recognition of foliar diseases. The description based on Dense SIFT achieved the best result with accuracy of 85%.