No abstract

No abstract

No abstract

Malaria is a disease that is threat to our life and it is caused by a parasite named protozoan. As we all studied that there are five species of malaria falciparum, p.vivax, p. malaria, p. ovale, and p. knowlesi. Some of the common symptoms of malaria are fever, fatigue, headache, cough, abdominal pain, nausea. Some of the causes of malaria are blood transfusion, maternal foetal transmission, by infected needle because of bite of female anopheles mosquito. Earlier number of method were used like malaria microscopy, RDT (Rapid Diagnostic Test). These tests were very costly so after their failure computer vision techniques came into limelight like image processing. Here microscopic images of blood samples were taken and using various techniques of image processing like preprocessing of an image, feature extraction, segmentation were used and were very popular in last few years but nowadays may machine learning classification models are used such as Naïve Bayesian classifier, Support vector machine classifier, K-nearest neighbor and many more and are very successful in predicting accuracy of detection of malaria parasite. In this paper we are going to detect malaria parasite and also which parasite it is with the help of a pre-processed image using region filling and then using canny edge detection for smoothening of edges in combination with watershed segmentation using distance transform and at last we will predict RMSE (Root Mean Square Error) using SVM classifier. Basically, in this paper we are trying to decrease the RMSE while detecting malaria parasite and we will predict the error using SVM classifier and also with Ensemble classifier and we will compare which one is better using regression analysis.