Paddy is the world’s largest and commonly cultivated food crop and is the staple food of more than 60 percent of the world’s population. To detect paddy seeds, a new feature set based on colour was created. Based on six colour feature sets, this approach has been applied to determine the strength of different feature models. With an in-house imaging system and analysed with neural networks, images of all seeds were acquired. The basic colour features and the new features derived from these basic features are the characteristics added as input for classification. For four paddy varieties, these characteristics were used to evaluate the detection model. Karjat-6, Karjat-2, Ratnagiri-4 and Ratnagiri-24, respectively. The Very Most For precise classification, a satisfactory feature was established from the colour characteristics. The performance of different function models has been investigated. Features based on method 3 were found to be bad, and better performance was shown by method 6, where the normalised values are subtracted from all three bands. For most of the classification cases, the performance was in bulk. This method 6, called as proposed-color2, achieved an 88.0 percent accuracy rate. Thus, colour moments have the ability to boost the performance of computer vision systems in terms of classification. The precision was lower since K6 and K2 have more or less the same colour, so the form and texture characteristics could also improve the accuracy. Become included.
Author (s) Details
Dr. Archana A. Chaugule
Department of Computer Engineering, PCCOE&R, Pune, India.
Dr. Suresh N. Mali
Dr. G H Raisoni College of Engineering and Management, Pune, India.
View Book :- https://bp.bookpi.org/index.php/bpi/catalog/book/349