Search for: All records

In recent years, the concentration of precious metals and hazardous pollutants in discarded consumer-grade computer enclosures has increased significantly, coinciding with e-waste generation in Asia reaching approximately 30 million tons annually. However, the high cost and low efficiency of manual disassembly present substantial obstacles to the effective recycling of such enclosures. Robotic disassembly has emerged as a promising alternative. To enable accurate acquisition of three-dimensional (3D) geometric data for robotic operations, we propose a 3D measurement method based on multi-color high dynamic range imaging. This method employs a seven-color illumination strategy and exploits the spectral response characteristics of a color camera to different wavelengths, effectively mitigating the reconstruction errors caused by overexposure on highly reflective surfaces—an issue common in traditional techniques. The proposed approach provides complete and reliable 3D morphological information to support robotic arm manipulation. Experimental results confirm that the method accurately captures the 3D profiles of reflective components such as CPUs and motherboards. Moreover, validation across computer enclosures of different brands and form factors demonstrates the method’s robustness and practical applicability in a wide range of e-waste disassembly scenarios. 

This study compares the accuracy of circular and linear fringe projection profilometry in the aspects of system calibration and 3D reconstruction. We introduce, what we believe to be, a novel calibration method and 3D reconstruction technique using circular and radial fringe patterns. Our approach is compared with the traditional linear phase-shifting method through several 2 × 2 experimental setups. Results indicate that our 3D reconstruction method surpasses the linear phase-shifting approach in performance, although calibration efficiency does not present a superior performance. Further analysis reveals that sensitivity and estimated phase error contribute to the relative underperformance in calibration. This paper offers insights into the potentials and limitations of circular fringe projection profilometry. 

This paper proposes a phase-to-depth deep learning model to repair shadow-induced errors for fringe projection profilometry (FPP). The model comprises two hourglass branches that extract information from texture images and phase maps and fuses the information from the two branches by concatenation and weights. The input of the proposed model contains texture images, masks, and unwrapped phase maps, and the ground truth is the depth map from CAD models. A loss function was chosen to consider image details and structural similarity. The training data contain 1200 samples in the verified virtual FPP system. After training, we conduct experiments on the virtual and real-world scanning data, and the results support the model’s effectiveness. The mean absolute error and the root mean squared error are 1.0279 mm and 1.1898 mm on the validation dataset. In addition, we analyze the influence of ambient light intensity on the model’s performance. Low ambient light limits the model’s performance as the model cannot extract valid information from the completely dark shadow regions in texture images. The contribution of each branch network is also investigated. Features from the texture-dominant branch are leveraged as guidance to remedy shadow-induced errors. Information from the phase-dominant branch network makes accurate predictions for the whole object. Our model provides a good reference for repairing shadow-induced errors in the FPP system.