Cropdiseasesareahugedangertofoodsecurity,butduetoalackofinfrastructureinmanyregionsoftheworld,earlyidentificationisdifficult.Plantdiseaseisapersistentproblemforsmall farmers, posing a risk to their livelihood and food security. Image categorization inagriculture has been achieved by the recent revolution in smartphone penetration andcomputer vision models. Convolutional Neural Networks (CNNs) are state-of-the-art inimage recognition and can deliver a quick and accurate diagnosis. A Convolutional NeuralNetwork(CNN)is a Deep Learning method thatcan take an imageas input,assignimportance (learnable weights and biases) to distinct aspects/objects in the image, anddistinguish one from the other. The ultimate objective was to get real time dataset for inputimagesthatwerecollectedoforange,cottonandsweetlimethesearetheplantandtressthataremostlygrowninVidarbhaalsothetypesofdiseasesthatmostlyaffectstheirgrowthwithnam and their occurrence ,to get efficient output with a decision of category for eachindividual pixel and for segmentation of pixels, we have used semantic segmentation whichlabel each pixel of an image with a corresponding class of what is being represented. So, inthispaperweareclassifyingthepixelsanddetectingthediseasedpartfirst,andonthatbasispredictingthediseaseintheleaf.