Abstract:Uilized PC vision and deep learning methods tochoose and plant sound billets, which expanded plant populace and the yield perhectare of sugarcane planting. We utilized notable convolutional neural network(CNN) structures to deal with enormous picture datasets and move learningmethods to extend the outcomes to various sugarcane assortments. It would beextremely tedious to gather and mark huge datasets for every sugarcaneassortment, for which quality investigation is required, preceding planting. Weutilized a two-venture move learning interaction to stretch out the prepareddesign to new assortments. We looked at results got during move learningutilizing AlexNet, VGG-16, GoogLeNet, ResNet101 structures to traditional PCvision techniques. Our objective was to decide the best way to deal withidentify harmed and great billets in the most brief preparing time. Best bringsabout both time and exactness were gotten with AlexNet. For AlexNet, we lookedat stages of three sugarcane assortments to track down the best model todistinguish the sound sugarcane billets. We at that point diminished thequantity of pictures utilized to retrain the model to decide tradeoff amongtime and execution. Eventually, one requirements a couple dozen billets of thenew assortment to retrain the network. Our methodology prompted significantaugmentations in the yield per hectare going from 33 to 80% contingent upon sugarcaneassortment
Keywords:—Agricultural robotics,computer vision, convolutional neural networks, sugarcane, transfer learning..