The complex components often require a variety of processes in the manufacturing process, such as turning, milling, grinding, polishing etc. Therefore, it is inevitable to produce defect features on the surface of the component. The defective surfaces will directly affect the performance of the entire component, so it must be identified during production and inspection. In this paper, based on the excellent curve feature recognition and sparse representation of curvelet transform, a defect extraction method based on the curvelet transform for feature separation in transform domain is proposed. The effectiveness of the method is proved by simulation results and experimental examples.
A type of distortion-free single-pixel ghost imaging without lenses is successfully realized by using a mobile phone. The proposed method is not sensitive to imaging distance. When the ratio of imaging distance to the size of the imaging object is close to 1 ∶ 2, clear and distortion-free imaging can still be achieved. This imaging entails low requirements for system hardware. A clear image can be realized when the brightness of a mobile phone is 70 lumens. Benefiting from the small volume, high portability, and wide application of mobile phones, as well as the broad application prospects of ghost imaging, the ghost imaging based on mobile phones can provide a portable and distortion-free imaging and detection method with significant applications in biological recognition and health monitoring.
With the development of precision optical engineering, higher manufacturing qualities are demanded for advanced optical systems. The characterization of the surface topographies of optical elements is required to be more specific and more comprehensive. In this paper, the contourlet transform is adopted to extract the topological features of optical elements. The performance of the contourlet transform(CT) is analyzed carefully. The multiscale analysis techniques based on contourlet transform for peak/pit extraction, tool trace identification and sharp edge detection on non-smooth microstructured optical surfaces were shown. The experimental examples are given to demonstrate the validity of the proposed method.