DATASET

Detecting disease with the help of fecal images is significantly a meaningful research. The diseases have the critical need for early and accurate detection to mitigate their impact. Traditional diagnostic methods often prove cumbersome, expensive, and reliant on specialized expertise. In response to that, this paper investigates the efficacy through a compared analysis of five prominent models: EfficientNetB7, VGG19, MobileNetV3, Vision Transformer, and Swin Transformer and acquired various accuracy outputs. Rather than relying on a single model's output, the majority voting algorithm in ensemble learning was employed achieving an accuracy rate of \99.25% resulting in improved performance and robustness.