This paper presents a method for detecting bleeding and tumors within capsule endoscopy (CE) images. Because CE can be used for visual, non-invasive examinations of the small bowel, it has recently become widely used. However, as a capsule progresses along the gastrointestinal tract, it collects vast quantities of images that make diagnosis a very time-consuming task. To address this problem, many computational approaches for anomaly detection have been proposed. Common to most approaches is the belief that color, texture, and shape are the most promising features for detecting anomalies within CE images. However, given that the requirements for each type of feature vary according to the anomaly, generally, it is essential to apply a complicated combination of techniques for multiple-feature extraction. In this study, in order to realize a scheme that covers the features of color, texture, and shape and can be applied to lesion areas of various sizes, a geometric image feature called local-contrast-enhanced higher-order local auto-correlation (LCE-HLAC) is proposed. Moreover, although the HSV color space is generally regarded as being appropriate for the analysis of CE images, imbalances in the distributions of utilized hue components limit discriminatory performance for normal and anomalous images. Accordingly, an image pre-processing method that uses a non-linear conversion model for the HSV color space is also proposed. Anomaly detection is implemented using a support vector machine classifier. The results of experiments, conducted with normal, bleeding, and tumor images obtained from 28 patients, demonstrate both the feasibility and superiority of the proposed method for both bleeding and tumor detection tasks.

Capsule endoscopy is a patient-friendly endoscopy broadly utilized in gastrointestinal examination. However, the efficacy of diagnosis is restricted by the large quantity of images. This paper presents a modified anomaly detection method, by which both known and unknown anomalies in capsule endoscopy images of small intestine are expected to be detected. To achieve this goal, this paper introduces feature extraction using a non-linear color conversion and Higher-order Local Auto Correlation (HLAC) Features, and makes use of image partition and subspace method for anomaly detection. Experiments are implemented among several major anomalies with combinations of proposed techniques. As the result, the proposed method achieved 91.7% and 100% detection accuracy for swelling and bleeding respectively, so that the effectiveness of proposed method is demonstrated.