Chin-Sheng Chen

A novel Fourier descriptor based image alignment algorithm for automatic optical inspection

This paper presents a Fourier descriptor based image alignment algorithm (FDBIA) for applications of automatic optical inspection (AOI) performed in real-time environment. It deliberates component detection and contour tracing algorithms and uses the magnitude and phase information of Fourier descriptors to establish correspondences between the target objects detected in the reference and the inspected images, so the parameters for aligning the two images can be estimated accordingly. To enhance the computational efficiency, the proposed component detection and contour tracing algorithms use the run length encoding (RLE) and Blobs tables to represent the pixel information in the regions of interest. The Fourier descriptors derived from the component boundaries are used to match the target objects. Finally, the transformation parameters for aligning the inspected image with the reference image are estimated based on a novel phase-shifted technique. Experimental results show that the proposed FDBIA algorithm sustains similar accuracy as achieved by the commercial software Easyfind against various rotation and translation conditions. Also, the computational time consumed by the FDBIA algorithm is significantly shorter than that by Easyfind.

An unconstrained virtual bone clamper for a knee surgical robot using visual servoing technique

Abstract Medical robotics has high potential to bring patients real benefits due to its high accuracy positioning capability. However a patients organ movement is still a challenge for medical robotics in achieving the expected high positioning accuracy. In orthorpaedic surgery, the target object is hard tissue. Obtaining accurate positioning is usually easier than with a soft tissue target. In the past, the solution was to use a mechanical clamper to fix and prevent the bone from moving. However, the invasive bone clamping approach has several drawbacks, such as additional damage to the bone, large errors for a loose bone, and constraints of ligament adjustment etc. Therefore, in this paper, an unconstrained virtual bone clamper has been proposed. Its aim is to provide the same bone freezing function as a traditional bone clamp, but it is implemented by software. The drawbacks of a physical bone clamp can be avoided. Experimental results on plastic sawbones resection demonstrate that using the virtual clamp, a robot can achieve accuracy as good ad when using a physical clamp whether the bone is fixed in place or moving.

An accelerating CPU based correlation-based image alignment for real-time automatic optical inspection

Display Omitted We propose an accelerating CPU based correlation-based image alignment.The image pyramid search scheme is combined with the parallel computation.Sub-pixel accuracy is used to attain the more accurate image alignment.It can align the accurate pose of the template image within the inspected image. This paper proposes an accelerating correlation-based image alignment using CPUs for time-critical applications in automatic optical inspection (AOI). In order to improve computation efficiency, the image pyramid search scheme is combined with the parallel computation. The image pyramid search scheme is employed first to quickly find certain objects in both monochrome and color images with rotation, translation and scaling. Sub-pixel accuracy is then used to attain the more accurate results at the sub-pixel level. In our experimental results, rotation accuracy is smaller than 0.218?, and the speed is increased between 277 and 20,841 times. According to translation, rotation and scaling tests, the errors of rotation, translation and scaling are 0.2?, 2.07pixel and 0.55%, respectively. These results show that the proposed method is suitable for dealing with correlation-based image alignment for time-critical applications in automatic optical inspection.

Intelligent Sliding-Mode Control for Knee Rehabilitation System

In this paper, we present the design and control of continuous passive motion (CPM) machine based on Constraint-Induced Movement Therapy (CIMT) for a patient who needs knee rehabilitation. First, the dynamic model of the CPM machine is derived by the principal of virtual work in dynamics. Then, an intelligent sliding-mode control (ISMC) system which involved recurrent Chebyshev neural network (RCNN) estimator to estimate the unknown external disturbance and uncertainty online is proposed to track the angular position and velocity of the CPM machine. Furthermore, we prove that the proposed ISMC system is asymptotically stable via Lyapunov theory. Finally, the experimental results are illustrated to demonstrate that the ISMC system can effectively reduce chattering phenomenon and precisely track reference trajectory of the CPM machine.

Fuzzy Neural Network Compensator for Gantry Stage Synchronous Motion

In this paper, a fuzzy neural network (FNN) compensator is proposed for the synchronous motion control of a gantry position stage. Firstly, the cascade control strategy is applied to reduce the single axis position tracking error. However, the synchronous error between dual servo systems is inevitable due to their inequality in characteristics and the environmental uncertainties. The FNN compensator and an online learning algorithm perform a fuzzy reasoning with two inputs of synchronous position and velocity errors between dual drive servo systems and generate the compensated force; the compensated force is fed back to the controller of each axis. The online learning algorithm adjusts the connected weighting of the neural network by using a supervised gradient descent methods, such that the define error function can be minimized. Finally, two kinds of position commands with high and low frequency are designed for the experiments, and the experimental results show that the proposed FNN compensator is feasible to improve the synchronous error of gantry stage.

IC Test Probe Measurement System

Integrated-Circuit (IC) test probes have been designed variously and manufactured flexibly cause of diverse IC packaged forms. To successfully test the functions of the packaged IC, the packaged industry should carefully select a suitable probe type based on the size or pitch of the ball and the lead of the packaged IC. One plunger of the probe is used to be connected with the packaged IC; the other plunger is connected to the testing devices. A probe basically is composite of a barrel and two plungers whose head might be the crown or the pierced type. The main diameter of the probe’s barrel varies from 0.2 millimeter to 0.6 millimeter; the total length of the probe could be from 1 millimeter to 7 millimeter. The probes ought to be detected before implanting them into the housing socket. However, the presenting probe measure method, eye checking by human beings, cannot fulfill the diverse types of probes because the size of IC becomes more delicate and smaller. This paper aims to develop an intelligent system that can detect and recognize more than ten items of a probe. The experimental results demonstrate that the proposed system can effectively measure the different probes.

A Survey of 2D and 3D Image Matching in Industrial Object Alignment

Abstract Image matching is a practical tool in the computer vision domain. 2D and 3D image matching can be based on (2D) features or on view and point (3D) which can be cloud based. Intensive practical research has been done on both 2D and 3D image matching and the methods are generally designed to meet the requirements of industrial alignment applications and include such aspects as full rotation, scaling and the handling of multiple objects. The aim of this study is the presentation of a review of 2D and 3D image matching and the provision of a comprehensive reference source for those involved in image matching. Graphical abstract

A Real-Time NCC-Based Template Matching on Modern CPUs

This paper proposes an optimal method for NCC-based template matching on modern CPUs for time-critical applications. In order to achieve the superior computation efficiency, the brand-and-bound (BB) scheme and the streaming SIMD extensions 2 (SSE2) instructions are employed to quickly find out the target object with rotation, translation and scaling in monochrome or color image. And we show how to reject unpromising image location very quickly using BB scheme in search process. Furthermore, an efficient implementation for similarity coefficient calculation is also pointed out by using the integration SSE2 instructions. Finally, the results show that the proposed method is very powerful when dealing with the NCC-based template matching in monochrome and color images.

A Novel Auto-Focus Measurement System for Mesh Membrane

In this paper, we develop the auto-focus measurement system which includes the hardware and software systems. The hardware system is composed of both modules of the motion stage and the auto-focus optic-mechanism. And the software system consists of two phases: (1) training the response surface model (RSM) for auto-focus technology in offline phase and (2) automatically capturing mesh images by RSM and robustly measuring holes by the area-based diameter (ABD) and average diameter (AD) in online phase. The experiments presented that the proposed auto-focus technology using RSM can effectively obtain the correct focusing position. The accuracy of the diameter on a hole can reach up at most 0.891 μm using the above two algorithms. The successful measure rate is almost 98 %; the measure time is averagely 4.8 seconds per mesh. The above results revealed that the proposed system is not only robust for the auto-focus technology but also efficient and effective for the mesh membrane measurement.

Field Distortion Compensation for Galvanometric Scanning Machines

The laser manufacture has been widely used in micro-machining as a burgeoning technology. Galvanometric scanning systems is an important component in laser machines, but the galvanometric scanning system is usually combines the image field distortion that is from the mechanism and optical devices. The paper has presented a novel correction algorithm based on ride regression method for compensates the image field distortion. And the Euclidean distance error and the maximum error are 0.0422 and 0.09, respectively. This novel correction technique can effectively increase the accuracy of laser spot position and can further improve the performance of the galvanometric scanning system.