Plant Leaf Disease (PLD) detection is helpful for several fields like Agriculture Institute and Biological Research. The country’s economic growth depends on the productivity of the agricultural field. Recently developed models based on deep learning give more accurate and precise results over the detection and classification of PLD while evolving through image processing approaches. Many image processing are used for the identification and classification of PLD. The quality of agricultural products is mainly affected by several factors like fungi, bacteria, and viruses. Thesefactors severely destroy the entire growth of the plant. Hence, some outperformed models are needed to detect and identify the severity level of plant diseases yet, the identification requires more time and has astruggle to identify theappropriate type of disease based on its symptoms. Therefore, several automatic detection and classification models are developedto avoid the time complexity. Computerized image processing approaches are utilized for crop protection, which analyzes the color information of leaves from the collected images. Hence, image processing techniques play an important role in the identification and classification of PLD. It gives more advantages by lowering the task of illustrating crops on large farms and detecting the leaf diseases at theinitial stage itself based on the symptoms of the plant leaves. While implementing a new model, there is a need to study various machine and deep learning-based structures for PLD detection approaches. This research work provides an overview of variousheuristic approaches, machine learning, and deep learning models for the detection and classification of PLD. This research work also covers the various constraints like PLD detection tools, performance measures, datasets used, and chronological review. Finally, the research work explores the research findings and also the research gaps with future scope.