In India, agriculture is the largest producer of The Gross Domestic Product (GDP). GDP significantly influences the economic conditions of any nation. Globally, the impact of plant diseases on agricultural productivity, causes major economic losses and food security problems. Plant diseases lead to a decrease in the quality and quantity of agricultural products. Cereal plant diseases are a major threat to food security. Early diagnosis of crop diseases using accurate and automated detection techniques can improve the quality of food production and reduce economic losses. Using a public dataset of 87,867 images divided into 38 healthy and diseased leaves collected under controlled conditions. This model is trained on a deep convolutional neural network to detect and classify many plant species. The proposed system can detect diseases of 14 crops such as apple, blueberry, acerola, corn, etc. Model performance is evaluated based on classification accuracy and mean score. A convolutional neural network has been involved in image classification. This model includes sequential layers, convolutional layers, and maximum voting layers. This model also completed a smoothing layer followed by a density layer using Relu activation for multi-class classification with Softmax activation. Finally, after the first dense layer, a Dropout Regularization with a rate of 0.25 is applied to avoid overfitting. The evaluation showed excellent global classification accuracy, with around 97% trained using Convolutional Neural Networks (CNN). This paper also highlights challenges and future directions to provide valuable information to researchers and practitioners involved in agricultural technology and the management of plant diseases.