A SYSTEM AND METHOD FOR CLASSIFYING SNAKE IMAGES USING CONVOLUTIONAL NEURAL NETWORK

• Main Authors: Chaduvula, Kavitha, Markapudi, Babu Rao, Srinivas, Kalyanapu, Srinivasarao, Tumma
• Format: Patent
• Language: English
• Published: 09.09.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

A SYSTEM AND METHOD FOR CLASSIFYING SNAKE IMAGES USING CONVOLUTIONAL NEURAL NETWORK The present invention discloses a system and a method for classifying snake images using 5 convolutional neural network (CNN). The system is comprised of an image capturing source (005), plurality of convolutional layers (001,003), a max pooling layer (002), a fully connected layer (004) and a graphical processing unit to classify the images of snakes such as cobra and mamba. The method involves capturing the image (102), filtering (104, 108) of the input data/image by multiple convolutional layer. The pooling layer's output (106) is 10 flattened and complied (110) on being passed through a fully connected layer. The neural network's output recognises and distinguishes images of snakes such as cobras and mambas in images (112). Figure 1 shall be reference figure. - 16- Application No: Applicant Name: Total No of Sheets: 4 Page 1 of 4 102 Capturing the image Filtering through first convolutional layer 106 Minimizing the number of parameters through the max pooling layer 108 Filtering through second convolutional layer 110 Flattening and compiling through a fully connected layer Classification of image into either Cobra or mamba Figure 1 Flow chart of the method for classifying snake images