• R Anitha
• N Juliet
• K Malathy
• R Rajeshwari

Wireless capsule endoscopy (WCE) is a technique that is increasingly being utilized to visualize the gastrointestinal (GI) tract. Capsule endoscopy exams are typically suggested as an additional monitoring mechanism and can aid in the detection of polyps, bleeding, and other abnormalities. Automatic image processing, computer vision, and learning algorithms are required to assess the vast volume of video data generated by WCE examinations. Recently, methods for automatic polyp detection have been presented, with varying degrees of effectiveness. While bleeding detection in colonoscopy and other standard endoscopic procedure-based images is maturing, detecting bleeding automatically in WCE is a difficult problem due to its unique imaging properties. This article takes a deep neural network method and offers a model called ABID (Automated Bleeding Image Detection) for classifying WCE bleedy images. It combines a color feature extraction layer with a fully convolutional network (FCN) model. ABID initially computes using the FCN model due to its reduced computational power demand, and then feeds the output to the CNN for further processing. A dataset of 1550 WCE images is used to train and test the model. The performance of the ABID is evaluated using the accuracy, precision, recall, and F1 score. The results reveal that the accuracy, precision, recall, and F1 score obtained were 94.51, 98.3, 94, and 97, respectively.

Wireless Capsule Endoscopy, Image Segmentation, Image Pre-Processing, CNN, Polyp Detection, Deep Learning.

"ABID (Automated Bleeding Image Detection) Using fully convolutional network (FCN) and Combined Texture and Color Features", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 7, page no.a110-a116, July-2022, Available :http://www.jetir.org/papers/JETIR2207014.pdf

"ABID (Automated Bleeding Image Detection) Using fully convolutional network (FCN) and Combined Texture and Color Features", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 7, page no. ppa110-a116, July-2022, Available at : http://www.jetir.org/papers/JETIR2207014.pdf