Skin diseases are among the most common health problems worldwide and are associated with considerable comorbidities. The objective and accurate quantification of the severity is essential to monitor the progress and healing of the disease, to observe the effects of medication, and to effective decision-making on the treatment method. Psoriasis is one of such diseases affecting 2-3% of the global population. To evaluate the degree of severity of the disease PASI index is widely used. This method assigns the single objective score ranging from 0-72, nonetheless, the individual score of each region is determined subjectively.
As a part of 3DHAP project, the purpose of the thesis is to design a stereo camera system, which is capable of differentiating between the induration levels defined by PASI. The stereo camera system has limited overlap, which means that only a limited area can be reconstructed. To increase the area of reconstruction error-state Kalman filter visual SLAM algorithm is investigated. These experiments are conducted on simulated data and real data. In the case of simulated data use of gyroscope is also investigated.
The initial prototype of the stereo camera has sufficient depth resolution to differentiate between the different levels of induration in accordance with PASI. The ES-EKF SLAM is also able to generalize the camera trajectory and reconstruct the object both on simulated as well as real data. Further, it is also seen that the ES-EKF SLAM with only visual information is not robust and additional sensors are required.
BlueJeans videoconference join information:
Meeting URL
https://bluejeans.com/550304744
Meeting ID
550 304 744
Want to dial in from a phone?
Dial one of the following numbers:
+31.20.808.2256 (Netherlands (Amsterdam))
(see all numbers - https://www.bluejeans.com/numbers)
Enter the meeting ID and passcode followed by #
