Plant diseases create one of the most serious risks to the world's food supply, reducing agricultural production and endangering millions of people's lives. These illnesses can destroy crops, disrupt food supply networks, and increase the danger of food deficiency, emphasizing the importance of establishing strong methods to protect the world's food sources. The approaches of deep learning have transformed the field of plant disease diagnosis, providing sophisticated and perfect solutions for early detection and management. However, a prevalent concern with deep learning models is their susceptibility to a lack of generalization and robustness when faced with novel crop and disease categories that were not included in the training dataset. To tackle this problem, this study present a novel deep learning-based model that can differentiate between healthy and diseased leaves in various crops, even if the model was not trained on them. The main idea is to identify the diseased small leaf regions instead of the diseased leaf's overall appearance and to calculate the disease's prevalence rate on the entire leaf. To achieve efficient classification and to take advantage of the Inception model's superiority in disease recognition, this study use a small Inception model architecture, which can process small regions without sacrificing performance. This study trained and evaluated the proposed approach using the highly regarded PlantVillage dataset, which is the most used dataset because of its extensive and varied coverage, in order to verify its efficacy. The accuracy percentage of the proposed approach was 99.75\%. This novel method tackles the crucial problem of model generalization to a variety of crops and diseases in addition to improving the precision of plant disease diagnosis. Furthermore, it demonstrated its potential for wide applicability and support to global food security programs by outperforming the current methodologies in its capacity to identify any illness across any crop type.