Fig. 3From: Tile-based microscopic image processing for malaria screening using a deep learning approachSample visualization results of best-performing model (YOLOV4-tiny). The top row (a–c) shows detection results for three test images using the proposed method and the bottom row (d–f) shows detection results using the baseline approach. Ground truth bounding boxes are in green and predicted boxes are in red. The figure also shows how the images vary in color and infection rateBack to article page