Image Classification for Monkey type Detection using Thirty six Fold Perpetuated Residual Network (TSFPRN)

Monkeys have a big impact on the ecology. They assist in the dispersal of seeds and plant pollination in addition to eating insects that could destroy crops. A vital resource for scholars, environmentalists, and the general public is the classification of monkeys inorder to save the endangered types. In this paper, a Thirty six-fold Perpetuated Residual Network (TSFPRN) model is proposed for conducting image super resolution for Monkey classification to take place efficiently. The model considered regenerates deep characteristics in 36 folds to obtain a super resolution image as a low-resolution image's produced in the monkey species dataset of monkey type images. The TSFPRN consists of Perpetuated Residual Blocks, i.e., Perpetuated Residual Block 1 (PRB1) for recovering deep features and up sampling the same and similarly Perpetuated Residual Block 2 (PRB2) but for down sampling instead. For super resolution factors 2,4,6, our model attained PSNR values of 31.942, 32.368, 34.102, SSIM values of 0.8478, 0.8991 and 0.9179 and classification accuracy values of 99.42, 98.88, 97.69 respectively.