Applied Mechanics and Materials Vols. 764-765

Abstract: In this study, a weight switching tri-state median filter (TSM) is proposed for removing impulse noise in both smooth and complicated regions of the dealing images. The switching scheme adopted large center weight while removing impulse noise of the complicate regions and switched to small center weight while dealing with the smooth regions. The smooth and complicated regions were classified according to the statistics information of the dealing image. The quad-tree decomposition was used to reduce classification procedures. Simulation results demonstrate that the proposed center weight switching scheme makes the TSM effective in both smooth and complicated regions of images with different noise ratio.

Abstract: With continuous advancement in science and technology, the image quality has entered an era of full-HD. This study developed a high-reliability image processing system platform, based on the FPGA platform. By using a high-reliability hardware platform development process, and with the aid of the simulation software, this study simulated the transmission integrity of the high-speed digital signals on the PCB. The proposed method was used to build a FPGA-based high-reliability image processing system platform. The implementation in this study, with the length of the Clock and DQS signal line of DDR2 being controlled within 555 mil, was discussed, and the errors were analyzed. The simulated value of the tDQSCK was 195.048 ps, the measured value was 215 ps, and the standard value of the JEDEC was less than 350 ps. Between the simulated value and the measured value, there was only an error of about 9.3%, which meets the reliability requirement. The length tolerance of the signal line laid was 38.5% better than the standard value of the JEDEC.

Abstract: 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.

Abstract: Five novel ideas are proposed in this paper to integrate an active RFID tag with a thermal convection angular accelerometer on a flexible substrate, thus the new device becomes a wireless sensor. The first innovative idea is that this device is without any movable parts, so it is very reliable. The second new idea is that it is made on a flexible substrate, such as plastic or polyimide, the thermal conductivity of the flexible substrate is much lower than silicon, and thus it can save more power and very useful for mobile operation. The third new idea is to apply xenon gas in the chamber to conduct the heat instead of the traditional carbon dioxide, so no oxidation and performance degradation effects will be produced on the heater and thermal sensors. The fourth new idea is to integrate with an active RFID tag on the same substrate, thus the device becomes a more useful wireless angular acceleration sensor. The final new idea is to apply a hemispherical chamber instead of the conventional rectangular one. From the simulation results one can see it is a good idea to apply a non-floating type angular accelerometer with a hemispherical chamber, the sensitivity performance is the best (395K/(rad/sec²)), and the response speed (693μs) is also comparable to the traditional one by using a floating type with a rectangular chamber (593μs). In summary, from the considerations of reliability, cost and performance, the non-floating type angular accelerometer with hemispherical chamber is a better choice.

Abstract: Generally after a cylindrical steel bar is made from hot forming process, its contour will look like ‘U’ or ‘wavy’ topography. The preformed steel bar after steelmaking process needs to remove the outside oxide layer by lathe or peeling machine (commonly known as stripping). As a consequence, the deformation of steel bar will cause subsequent problems, such as eccentric rotating of steel bar as the lathe is running. Although it is necessary to remove the oxide layer, the deformation will increase waste materials. To prevent the problems, a straightening machine is usually used to straighten the cylindrical steel bar before the pickling process. However, most current straightening machines cannot measure cylindrical steel bar contour on-line instantly. This study proposed a test method for measure the contour of a large cylindrical steel bar based on multi-line structured light and machine vision. Break line method were performed to measure the contour of the cylindrical steel bar. The experimental result shows that the measurement error is in the range of ±0.4 mm after calibration. The study result concludes that the proposed method can be applied to straightening machine for real-time online measurement to improve straightening efficiency.

Abstract: We propose a tool to calibrate the coefficient variables of the Steinhart–Hart equation, which are used in temperature measurement with a negative temperature coefficient thermistor. The previous method modifies the coefficient variables manually, but the proposed tool takes a measured temperature and automatically modifies the coefficient variable. The proposed calibration tool provides a graphical user interface program for the convenience of users. It is applied to the 4-point temperature measurement of polymerase chain reaction and has a degree of precision of ±0.1 °C in the temperature measurement evaluation

Abstract: In this study, we propose a new solution based on Adaboost algorithm and Back Propagation Network (BPN) of Neural Network (NN) combining local and global features with cascade architecture to detect human faces. We use Modified Census Transform (MCT) feature that belong to texture features and is less sensitive to illumination for local feature calculation. By this approach, it is not necessary to preprocess each sub-window of the image. For classification, we use the structure of hierarchical feature to control the number of features. With only MCT, it is easy to misjudge faces, and therefore in this work we include the brightness information of global features to eliminate the false positive regions. As a result, the proposed approach can have Detection Rate (DR) of 99%, false positives of only 11, and detection speed of 27.92 Frame Per Second (FPS).

Abstract: We employed a gold coated glass cover plate to fabricate dual compact free running fiber laser combs based LIDAR (light detection and ranging) system with tunable difference of repetition rates. Our study shows that the ranging precision is 0.4 μm and the non-ambiguity range is 2.1 m with averaging time of 0.4 s for the distance of 1.09 m by the time-of-flight method. The system with 1 kHz-high update rate and the large non-ambiguity range is potentially useful for manufacturing or machining application applied in absolute distance measurement.

Abstract: Pyroelectric infrared (PIR) sensors can detect the presence of human without the need to carry any device, which are widely used for human presence detection in home/office automation systems in order to improve energy efficiency. However, PIR detection is based on the movement of occupants. For occupancy detection, PIR sensors have inherent limitation when occupants remain relatively still. Multisensor fusion technology takes advantage of redundant, complementary, or more timely information from different modal sensors, which is considered an effective approach for solving the uncertainty and unreliability problems of sensing. In this paper, we proposed a simple multimodal sensor fusion algorithm, which is very suitable to be manipulated by the sensor nodes of wireless sensor networks. The inference algorithm was evaluated for the sensor detection accuracy and compared to the multisensor fusion using dynamic Bayesian networks. The experimental results showed that a detection accuracy of 97% in room occupancy can be achieved. The accuracy of occupancy detection is very close to that of the dynamic Bayesian networks.

Abstract: This paper integrated a laser triangulation probe with a two-dimensional image measurement system to measure the height of an object using the auto-focusing method. In the laser triangulation probe, the laser diode was used as the light source and one-dimensional position sensitive detector was adopted to receive the position of the laser spots to detect the height. The laser triangulation probe was used for small height measurement and the auto-focusing method was used for large height measurement. Two standard gauges with thicknesses of 8mm and 8.5mm were used to verify our proposed method. With height difference of 500μm, the experiment results showed the measuring error was within ±3μm and standard deviation was about 0.1μm.