workbench

workbench

結果 (繁體中文) 1:

工作台

結果 (繁體中文) 2:

The research extends [48] by focusing on image processing solutions for solid waste collection. The paper presents the process of capturing bin image with a camera embedded in the bin. Specifically, during capturing the bin image the position of the camera should be focus on getting a centralized image of the bin area. The captured image should be further processed in order to correctly estimate the waste capacity in the bin. The research uses DTW for detecting and cropping the bin area; while Gabor Wavelet (GW) is incorporated for feature extraction of the bin image. The features extracted are then used to train a MLP classifier which is adopted to classify the bin level and perform estimations about the capacity of the waste in the bin. The classifier performance is evaluated with Receiver Operating Characteristic (ROC) curves and proved to be efficient with respect to the accuracy (i.e., level estimation of 98.5 percent). In [50], the authors propose an efficient waste collection model incorporating shortest path semi-static and dynamic routing.

結果 (繁體中文) 3:

工作臺