A novel feature based algorithm for soil type classification

Abstract Agriculture is the backbone of Bangladesh’s economy and it one largest employment sectors. In Bangladesh, population increasing rapidly at same time, total cultivable land decreasing significantly. To ensure maximum crop production using limited resources, essential to identify select appropriate type soil because different crops need types. Currently, there are two types methods available determine type, namely chemical image analysis. Although first accurate, expensive time consuming. On other hand, based classification much cheaper faster but its accuracy level low. this study, we present a novel feature algorithm that combines quartile histogram oriented gradients (Q-HOG), most frequent $$\varphi $$ φ -Pixels new selection method for classifying We have used four machine learning algorithms evaluated performance with sets features. also compared our work prominent recent works on image-based systems. The experimental results show proposed in terms standard evaluation metrics, accuracy, precision, F1_score, recall scores higher than existing