Enhance Support Vector Machine models for Cassava leaf disease variants identification and classification

Cassava is one of the two main staple crops grown and consumed in large quantities in Nigeria and despite its usefulness and benefits to mankind, cassava is susceptible to various diseases such as “cassava mosaic disease (CMD)”, “cassava green mottle disease (CGMD)”, “cassava bacterial blight disease (CBBD)”, and “cassava brown streak disease (CBSD)”. It is very tedious and inaccurate to identify cassava disease on its plant leaves physically, but failure to detect the disease on time affects both the quality and quantity of the product. Therefore, many researchers have developed different classification models based on machine learning techniques to detect disease on plant leaves, but many of these models are easily influenced by imbalanced datasets, the selection of irrelevant features, and fine-tuning of the hyperparameters in the classifier. Therefore, this study uses an equal number of datasets for the diseased dataset, and hybridises Binary Particle Swarm Optimisation (BPSO) with Reptile Search Algorithm (RSA) to fine-tune and select discriminating hyperparameters in the Support Vector Machine (SVM). A classification model (i.e., BPSO-RSA-SVM) was developed, trained and tested with the datasets. The results were compared with other state-of-the-art models. It was found that the BPSO-RSA-SVM achieved an accuracy of 96.73% compared with both BPSO-SVM and RSA-SVM models that achieved an accuracy of 95.51% and 94.25%, respectively. These results affirmed that hybridising two or more optimisation techniques will improve the performance of the classification model. Therefore, it is recommended that the model (BPSO-RSA-SVM) be used to detect other plant diseases on plant leaves.