Real-time segmentation of surgical instruments inside the abdominal cavity using a joint hue saturation color feature

A real-time automatic instrument tracking system on cataract surgery videos for dexterity assessment

“…There are other moving objects in the scene that have significant interactions with the instruments and such as the new lens and the old lens fragments. There has been limited previous work on localization and tracking of surgical instruments through image and video analysis, but in most of these cases the authors usually only deal with high quality images and very predicable patterns, or with virtual reality based surgery simulators [3], [4] and [5]. The developed system in this paper which is based on a robust multilevel tracking methodology, deals with the poor video quality and with the random characteristics of instruments.…”

Pose Estimation and Feature Tracking for Robot Assisted Surgery with Medical Imaging

“…These methods are based on the structure, mainly the apparent lines of the instrument [1,6], or on its frequential features [8,41]. In order to make the detection more robust and accurate, instruments can be marked with structuring markers as described before ( [1,43,22]) or frequential (color) markers ( [42,38]).…”

“…1.6) leading to grey regions with many specularities in the image. Prior researchs involving such endoscopic images have been conducted in the field of color image segmentation [8]. Once regions have been segmented, the region boundaries are ordered and used to perform a degenerate conic-based contours fitting.…”

Learning features on robotic surgical tools

“…Tool tracking has been addressed using various approaches, ranging from marker-based [5] and pixel segmentation [6] concepts to model-based [7,8] or template matching [9] algorithms. A recent approach combined multiple features with learning at the pixel level [10], however few, if any, address interest regions using rich feature descriptors on the tip of the tool.…”