Finding flaws in agricultural foodstuffs, especially potatoes, requires the use of machine visionVision and image processingImage processing techniques. The use of image processingImage processing and artificial intelligenceArtificial intelligence in agricultureAgriculture for the classification and identification of plant and fruit pests and illnesses has grown, and work in this area is still ongoing. Numerous diseases, such as early blight and late blight, have a significant impact on the quality and yield of potatoes, and manual clarification of these leaf diseases is time-consuming and labour-intensive. Effective and automated diagnosis of these diseases during the budding phase can help to improve the productivity of the potato crop despite the high degree of required. Various models have been put up in the past to identify various plant diseases. With the help of dubbed TuberNetTuberNet, this study seeks to offer a practical solution for identifying the numerous potato illnesses. More specifically, the study uses the TuberNetTuberNet to identify various potato leaf abnormalities utilising a strategy. To focus on the areas and improve the tactic’s capacity to recognise various illnesses, a spatial-channel attention technique is employed. Transfer learningTransfer learning is utilised to address the capacity. The model is put to the test on the open-source, difficult dataset known as PlantVillage, which contains pictures collected under a variety of hard backdrop settings, such as varying lightning numerous colour changes in leaves. Through the executed trials, the model has verified that our method is efficient for classifying potato diseases and can successfully handle distorted tasters. So, utilising the suggested model tool, farmers can both save money and increase their crop.