Wireless Capsule Endoscopy (WCE) is a non-invasive medical process that permits the assessment of the whole gastrointestinal tract that includes small intestine parts ahead of the conventional endoscope scope. This in turn needs the approach of a computer-aided scheme to assess the video frames for reducing the time of diagnosis. In this approach, the performance comparison on various existing techniques like color-based feature extraction, histogram-based approach, Discrete Wavelet Transform (DWT), K-nearest neighbor (KNN), K-means, and Support Vector Machine (SVM) techniques employed in the detection of bleeding to that of the proposed Gaussian Mixture Model (GMM) super-pixel model has been carried out. The study is carried out in terms of feature extraction models used so far and the classification approaches employed so far. Then the experimental study is carried out in terms of existing techniques and the performance is compared with GMM super-pixel feature extraction model and linear SVM to prove the effectiveness of the super-pixel-based model for bleeding detection. From the experimental analysis, the GMM super-pixel model is concluded better for automated bleeding detection.