Regression Model for Measurement of Wound Dimensions by Webcam Scanners and Time-of-Flight Sensors

One use of image processing is for medical equipment such as wound identification. This technology is carried out non-invasively by taking images so as to avoid direct touch with the wound thereby reducing the possibility of infection. The images obtained using the RGB camera will be used for color segmentation which will measure the wound dimensions. However, the image data is in the form of a raster, the distance will affect the pixel size. Therefore, it is necessary to consider the distance of the camera measurement to the object. The time-of-flight (ToF) method with a lidar sensor is used to calculate the distance of the camera to the object. It is necessary to calculate the ratio of the distance to the number of pixels obtained so that the value is always consistent. This study analyzed the use of appropriate ratios and regression systems on a webcam and a lidar sensor for measuring wound dimensions. The results of the study show that there is a regression model with a second-order polynomial relationship for the distance and number of pixels obtained consistently with an error value of less than 5% which shows very good results