In this study, we propose a deep learning-based method for large-area inspection aimed at the high-speed detection of micro hole diameters. Micro holes are detected and stored in large images using YOLOv8, an object detection model. A super-resolution technique utilizing ESRGAN, an adversarial neural network, is applied to images of small micro holes, enhancing them to high resolution before measuring their diameters through image processing. When comparing the diameters measured after 8x super-resolution with the results from existing inspection equipment, the average error rate is remarkably low at 0.504%. The time taken to measure an image of one micro hole is 0.470 seconds, which is ten times faster than previous inspection methods. These results can significantly contribute to high-speed measurement and quality improvement through deep learning.

The process of tattooing to mark the position of lesions in the colon is one of important functions of the conventional endoscope. However, commercial capsule endoscope (CE) devices cannot perform the tattooing procedure because they cannot accommodate the size of the tattooing device. In this paper, we propose a compact tattooing mechanism design which can be accommodated inside the CE. Two conical springs, two triggering modules and a needle that can be installed inside a volume of 840 mm3 are employed to perform the needle insertion/withdrawal and inject the ink. A triggering module to deploy the conical springs is designed to be activated by heating a Ni-Cr wire and melting Wood’s metal. In this study, the activation time of the triggering module is investigated based on a Wood’s metal heating simulation. In order to determine the proper conical springs to ensure the activation of the tattooing mechanism, the elastic force correlation between two conical springs is studied. Then, the components of the proposed tattooing mechanism are fabricated and assembled, and an ex-vivo test is performed. Conclusively, the proposed tattooing mechanism implements the correct needle stroke and the proper ink injection into the submucosal layer of a porcine colon.