Survey on Hybrid Techniques in The Classification of Nutrient Deficiency Levels in Citrus Leaves

there are many challenges when applying image processing technology and learning algorithms for plant nutrition detection. The proposed algorithms should always be very accurate and the margin of error should be very slight since a wrong detection could be very destructive to agricultural production. After detecting the presence of nutrient deficiencies in citrus leaves, further steps are needed to classify the types of nutrients that are deficient in citrus plants. Artificial Neural Network (ANN) and Support Vector Machine (SVM) methods are effective methods in artificial intelligence techniques for image classification. But both of them have weaknesses that can be improved by optimization techniques. After survey the previous research, this paper will proposed the design / methods of optimization ANN and SVM algorithms using Salp Swarm Algorithm (SSA). SSA is used to activate and searching the hidden layers on ANN. The value gain at the SVM process will be optimized by SSA to optimize the threshold using SSA methods.