An Implementation of Malaria Detection Using Regional Descriptor and PSO-SVM Classifier

TLDR: A new tool to diagnose malaria using regional descriptor and PSO-SVM classifier is made, which pathologists can easily detect malaria parasites, and they can achieve 98% accuracy.

Abstract: Malaria is a serious worldwide health issue which causes an expected 13,444 individuals in danger of malaria in 2017. The estimated cost of detection of malaria in India is 11,640 crores per year. So there is an urgent need for a new tool to diagnose malaria. Malaria is completely preventable and treatable disease. In this project, we make a new tool to diagnose malaria using regional descriptor and PSO-SVM classifier. The proposed work used various image processing techniques like image acquisition, image pre-processing, image segmentation, feature extraction and classification. The implementation work is mainly focusing on detection accuracy, computational time, less estimation time for parasite detection. In this way, the new tool for the detection of malaria parasites gives faster and accurate results, and by using this proposed methods pathologists can easily detect malaria parasites, and they can achieve 98% accuracy. This new tool is useful to reduce deaths.