Recent innovations in sensor technology enable manufacturers to distribute redundant sensors in manufacturing processes for quality monitoring, defect detection, and fault diagnosis. Even if a single sensor is relatively reliable, the large number of sensors in a distributed sensor system confronts us the almost unavoidable possibility that some of the sensors may malfunction. Without isolating sensor anomalies from the underlying process changes, abnormal sensor readings can cause frequent false alarms and jeopardize productivity. Traditionally, sensor system reliability has been ensured by employing off-line gage Repeatability and Reproducibility (R&R) calibration. But this off-line approach can be time consuming and costly for in-process distributed sensor systems. This paper will present a robust estimation procedure that automatically identify the observations related to suspected sensor failures. We first identify sensor redundancy and introduce an existing algorithm to assess the redundant level. We further suggest a decomposition technique, which helps to substantially reduce the computation expense of the existing algorithms for a large sensor system. Finally, the concept and procedure is illustrated using a distributed coordinate sensor system in a multi-station manufacturing system.

Sensors are increasingly embedded in manufacturing systems and wirelessly networked to monitor and manage operations ranging from process and inventory control to tracking equipment and even post-manufacturing product monitoring. In building such sensor networks, a critical issue is the limited and hard to replenish energy in the devices involved. Dynamic voltage scaling is a technique that controls the operating voltage of a processor to provide desired performance while conserving energy and prolonging the overall network's lifetime. We consider such power-limited devices processing time-critical tasks which are non-preemptive, aperiodic and have uncertain arrival times. We treat voltage scaling as a dynamic optimization problem whose objective is to minimize energy consumption subject to hard or soft real-time execution constraints. In the case of hard constraints, we build on prior work (which engages a voltage scaling controller at task completion times) by developing an intra-task controller that acts at all arrival times of incoming tasks. We show that this optimization problem can be decomposed into two simpler ones whose solution leads to an algorithm that does not actually require solving any nonlinear programming problems. In the case of soft constraints, this decomposition must be partly relaxed, but it still leads to a scalable (linear in the number of tasks) algorithm. Simulation results are provided to illustrate performance improvements in systems with intra-task controllers compared to uncontrolled systems or those using inter-task control.

Discrete wavelet transform has become a widely used feature extraction tool in pattern recognition and pattern classification applications. However, using all wavelet coefficients as features is not desirable in most applications -- the enormity of data and irrelevant wavelet coefficients may adversely affect the performance. Therefore, this paper presents a novel feature extraction method based on discrete wavelet transform. In this method, Shannon's entropy measure is used for identifying competent wavelet coefficients. The features are formed by calculating the energy of coefficients clustered around the competent clusters. The method is applied to the lung sound classification problem. The experimental results show that the new method performs better than a well-known feature extraction method that is known to give the best results for lung sound classification problem.

This paper presents a systematic approach to the design and implementation of an energy-efficient multi-sensor network. The nodes of the sensor network form the basis of a sectioned Bayesian network that can be used to determine the state of the system being monitored. A key issue in the design of Bayesian networks for monitoring engineering systems is to ensure that reliable inference scheme about the health state of the system can be made by combining information acquired from each sensor in the system into a single Bayesian network. As the size of the network increases, aggregating information made by all the sensors becomes computationally intractable. Hence, sectioning of the Bayesian network based on functional or logical constraints allows computational efficiency in aggregating information and reduces overall communication requirements. Furthermore, an in-network data processing scheme, motivated by the concept of Dynamic Voltage Scheduling, has been investigated to minimize computation energy consumption through dynamically adjusting the voltage supply and clock frequency of the individual sensors. As a result, the processor idle time can be better utilized for prolonged computation latency, leading to significantly reduced energy cost and increased computational efficiency.

We consider wireless sensor networks with multiple gateways and multiple classes of traffic carrying data generated by different sensory inputs. The objective is to devise joint routing, power control and transmission scheduling policies in order to gather data in the most efficient manner while respecting the needs of different sensing tasks (fairness). We formulate the problem as maximizing the utility of transmissions subject to explicit fairness constraints. We propose an efficient decomposition algorithm drawing upon large-scale decomposition ideas in mathematical programming, and prove the convergence of the algorithm. Furthermore, by exploring the property of the subproblem, we find that the utility maximization problem we consider can, in principle, be solved in polynomial time.

There is a missing link between a virtual development environment (e.g., a CAD/CAM driven offline robotic programming) and production requirements of the actual robotic workcell. Simulated robot path planning and generation of pick-and-place coordinate points will not exactly coincide with the robot performance due to lack of consideration in variations in individual robot repeatability and thermal expansion of robot linkages. This is especially important when robots are controlled and programmed remotely (e.g., through Internet or Ethernet) since remote users have no physical contact with robotic systems. Using the current technology in Internet-based manufacturing that is limited to a web camera for live image transfer has been a significant challenge for the robot task performance. Consequently, the calibration and accuracy quantification of robot critical to precision assembly have to be performed on-site and the verification of robot positioning accuracy cannot be ascertained remotely. In worst case, the remote users have to assume the robot performance envelope provided by the manufacturers, which may causes a potentially serious hazard for system crash and damage to the parts and robot arms. Currently, there is no reliable methodology for remotely calibrating the robot performance. The objective of this research is, therefore, to advance the current state-of-the-art in Internet-based control and monitoring technology, with a specific aim in the accuracy calibration of micro precision robotic system for the development of a novel methodology utilizing Ethernet-based smart image sensors and other advanced precision sensory control network.

One of the major problems of modern industrial robots is a lack of reliable perceptual systems that are similar to human vision in its abilities to understand visual scene and detect and unambiguously identify objects. The traditional linear bottom-up "segmentation-grouping-learning-recognition" approach to image processing and analysis cannot provide a reliable separation of an object from its background or clutter, while human vision unambiguously solves this problem. The modern computer vision can only recognize certain features from visual information, and it plays an auxiliary role, helping to build or choose appropriate 3-dimensional models of objects and visual scene. As result, designers of robotics systems must create for industrial robots artificial environments, which allowing for precise computations of 3-dimensional models within such environments. However, outside of such an artificial environment, the robot is dysfunctional. Biologically-inspired Network-Symbolic models do not compute precise 3-dimensional models, but convert image information into an "understandable" Network-Symbolic format, which is similar to relational knowledge models. Feature, symbol, and predicate are equivalent in the Network-Symbolic systems. A linking mechanism binds these features or symbols into coherent structures, and image converts from a "raster" into a "vector" representation that can be better interpreted by higher-level knowledge structures. Logic of visual scenes can be captured in the Network-Symbolic models and used for the disambiguation of visual information.

This paper shows the potential of using robotics for data acquisition within full-scale field trials. Robotics ensured simultaneously measurements from several sensors from GPS targeted sampling points. This was demonstrated by supporting a project developing methods to measure gap fraction and canopy structure in cereals. The project required measurements from ordinary barley canopy areas using a high- dynamic-range RGB camera, and a multi-spectral Cropscan radiometer. Further, the RGB camera required images from 12 different angles relative to the rows. In fulfilling these demands an existing robotic platform at Research Center Bygholm, Denmark, was enhanced. A payload software system was developed ensuring a simple and efficient interface between the robotic platform and the multiple sensor systems. The software of the Cropscan Multispectral Radiometer System was also altered to support remote control by the payload software. The robotic system collected data from a full scale field at two occasions.

The objective of this research is to extend the sensing capabilities of a multi-vehicle ground system by incorporating the environmental perception abilities of unmanned aerial vehicles. The aerial vehicle used in this research is a Miniature Aircraft Gas Xcell RC helicopter. It is outfitted with a sensor payload containing stereo vision cameras, GPS, and a digital compass. Geo- referenced images are gathered using the above sensors that are used in this research to create a map of the operating region. The ground vehicle used in this research is an automated Suzuki Mini-Quad ATV. It has the following onboard sensors: single-vision camera, laser range device, digital compass, GPS, and an encoder. The ground vehicle uses the above sensors and the map provided by the helicopter to traverse the region, locate, and isolate simulated land mines. The base station consists of a laptop that provides a communication link between the aerial and ground vehicle systems. It also provides the operator with system operation information and statistics. All communication between the vehicles and the base station is performed using JAUS (Joint Architecture for Unmanned Systems) messages. The JAUS architecture is employed as a means to organize inter-vehicle and intra-vehicle communication and system component hierarchy. The purpose of JAUS is to provide interoperability between various unmanned systems and subsystems for both military and commercial applications. JAUS seeks to achieve this through the development of functionally cohesive building blocks called components whose interface messages are clearly defined. The JAUS architecture allows for a layered control strategy which has specific message sets for each layer of control. Implementation of the JAUS architecture allows for ease of software development for a multi- vehicle system. This experiment demonstrates how an air-ground vehicle system can be used to cooperatively locate and dispose of simulated mines.

The prudent selection of the sampling points ensures that representative points to typify a feature surface are obtained. The rationale is that the larger the number of sample points, the better the estimate of the surface. Large samples however lead to large measurement times and consequently time-induced errors. It is believed that a priori knowledge of process-induced errors can help in minimizing the total number of sampling points. Modeling the initial points for search, or approximate locations of errors is the key to minimizing the sampling effort. Suitable search methodology can then be used to determine the actual location of errors. If the regression surface describing the actually measured points can be identified, an adaptive search can be conducted. To do this we are using a kind of function learning machines that has been extensively developed the last decade, the Support Vector Regression (SVR). This paper describes in general terms our methodology.

Presently there are no suitable non-invasive methods for precisely detecting the subsurface defects in logs in real time. Internal defects such as knots, decays, and embedded metals are of greatest concern for lumber production. While defects such as knots and decays (rots) are of major concern related to productivity and yield of high value wood products, embedded metals can damage the saw blade and significantly increase the down time and maintenance costs of saw mills. Currently, a large number of logs end up being discarded by saw mills, or result in low value wood products since they include defects. Nondestructive scanning of logs using techniques such as Ground Penetrating Radar (GPR) prior to sawing can greatly increase the productivity and yield of high value lumber. In this research, the GPR scanned data has been analyzed to differentiate the defective part of the wooden log from the good part. The location and size of the defect has been found in the GPR scanned data using the MATLAB algorithm. The output of this algorithm can be used as an input for generating operating instructions for a CNC sawing machine. This paper explains the advantages of the GPR technique, experimental setup and parameters used, data processing using RADAN software for detection of subsurface defects in logs, GPR data processing and analysis using MATLAB algorithm for automated defect detection, and comparison of results between the two processing methods. The results show that GPR in conjunction with the proposed algorithm provides a very promising technique for future on-line implementation in saw mills.

During the past several years, many new biological and chemical sensors have been or are being developed for infrastructure and environment protection, such as protecting water, indoor and outdoor air quality. However, there is a lack of fundamental system level research that develops the methodologies to optimize such a sensor network to maximize the protection and minimize the system cost. This paper describes a preliminary study to address the above questions. In this study, the evaluation criteria for a sensor system that is used to protect a building from airborne hazards are identified. Common building attack scenarios are described and simulated for a small commercial building. Genetic Algorithm is applied for each attack scenario to optimize the sensor sensitivity, location, and amount to achieve the best system behavior while reduce the total system cost. Assuming that each attack scenario has the same occurrence possibility, optimal system designs that present the best behavior for all attacking scenario are obtained.

This paper describes the simulation and on-line fuzzy logic control of tube bending. By combining elasticity and plasticity theories, a conventional model was developed. The results from simulation were compared with those obtained from testing. The experimental data reveal that there exists certain level of uncertainty and nonlinearity in tube bending, and its variation could be significant. To overcome this, a on-line fuzzy logic controller with self-tuning capabilities was designed. The advantages of this on-line system are (1) its computational requirement is simple in comparison with more algorithmic-based controllers, and (2) the system does not need prior knowledge of material characteristics. The device includes an AC motor, a servo controller, a forming mechanism, a 3D optical sensor, and a microprocessor. This automated bending machine adopts primary and secondary errors between the actual response and desired output to conduct on-line rule reasoning. Results from testing show that the spring back angle can be effectively compensated by the self- tuning fuzzy system in a real-time fashion.

Multispectral sensors produce images with a few relatively broad wavelength bands. Hyperspectral remote sensors, on the other hand, collect image data simultaneously in dozens or hundreds of narrow and adjacent spectral bands. These measurements make it possible to derive a continuous spectrum for each image cell, generating an image cube across multiple spectral components. Hyperspectral imaging has sound applications in a variety of areas such as mineral exploration, hazardous waste remediation, mapping habitat, invasive vegetation, eco system monitoring, hazardous gas detection, mineral detection, soil degradation, and climate change. This image has a strong potential for transforming the imaging paradigms associated with several design and manufacturing processes. In this paper, we describe a novel approach for fast indexing of multi-dimensional hyperspectral image data, especially for data mining applications. The index exploits the spectral and spatial relationships embedded in these image sets. The index will be employed for knowledge retrieval applications that require fast information interpretation approaches. The index can also be deployed in real-time mission-critical domains, as it is shown to exhibit speed with high degrees of dimensionality associated with the data. The strength of this index in terms of degree of false dismissals and false alarms will also be demonstrated. The paper will highlight some common applications of this imaging computational paradigm and will conclude with directions for future improvement and investigation.

Prolog offers a very different style of programming compared to conventional languages; it can define object properties and abstract relationships in a way that Java, C, C++, etc. find awkward. In an accompanying paper, the authors describe how a distributed web-based vision systems can be built using elements that may even be located on different continents. One particular system of this general type is described here. The top-level controller is a Prolog program, which operates one, or more, image processing engines. This type of function is natural to Prolog, since it is able to reason logically using symbolic (non-numeric) data. Although Prolog is not suitable for programming image processing functions directly, it is ideal for analysing the results derived by an image processor. This article describes the implementation of two systems, in which a Prolog program controls several image processing engines, a simple robot, a pneumatic pick-and-place arm), LED illumination modules and a various mains-powered devices.

The current trends in industry include integration of an information and knowledge base network with a manufacturing system, which coined a new term, E-Manufacturing. From the perspective of E-Manufacturing, any production equipment and its control functions do not exist alone, but become a part of the holistic operation system with distant monitoring and fault diagnostic capabilities. The key to this new paradigm is the accessibility to a remotely located system and having the means of responding to a changing environment. In this study, a new methodology in predicting a system output has been investigated by applying a data mining technique and a hybrid type II fuzzy system in CNC turning operations. The purpose was to generate a supplemental control function under the dynamic machining environment, where unforeseeable changes may occur frequently. Two different types of membership functions were developed for the fuzzy logic systems and also by combining the two types, a hybrid system was generated. Genetic algorithm was used for fuzzy adaptation in the control system. Fuzzy rules are automatically modified in the process of genetic algorithm training. The computational results showed that the hybrid system with a genetic adaptation generated a far better accuracy. The hybrid fuzzy system with genetic algorithm training demonstrated more effective prediction capability and a strong potential for the implementation into existing control functions.

Benefits of laser ignition over the electrical ignition system for Compressed Natural Gas (CNG) engines have fuelled automobile industry and led to an extensive research on basic characteristics to switch over to the emerging technologies. This study was undertaken to determine the electrical and physical characteristics of the electric spark ignition of single cylinder petrol/CNG engine to determine minimum ignition requirements and timeline of ignition events to use in subsequent laser ignition study. This communication briefly reviews the ongoing research activities and reports the results of this experimental study. The premixed petrol and CNG mixtures were tested for variation of current and voltage characteristics of the spark with speed of engine. The current magnitude of discharge circuit was found to vary linearly over a wide range of speed but the stroke to stroke fire time was found to vary nonlinearly. The DC voltage profiles were observed to fluctuate randomly during ignition process and staying constant in rest of the combustion cycle. Fire to fire peaks of current amplitudes fluctuated up to 10% of the peak values at constant speed but increased almost linearly with increase in speed. Technical barriers of laser ignition related to threshold minimum ignition energy, inter-pulse durations and firing sequence are discussed. Present findings provide a basic initiative and background information for designing suitable timeline algorithms for laser ignited leaner direct injected CNG engines.

Economic considerations of laser induced ignition over the normal electrical ignition of direct injected Compressed Natural Gas (CNG) engines has motivated automobile industry to go for extensive research on basic characteristics of leaner unconventional fuel mixtures to evaluate practical possibility of switching over to the emerging technologies. This paper briefly reviews the ongoing research activities on minimum ignition energy and power requirements of natural gas fuels and reports results of present laser air/CNG mixture absorption coefficient study. This study was arranged to determine the thermo-optical characteristics of high air/fuel ratio mixtures using laser techniques. We measured the absorption coefficient using four lasers of multiple wavelengths over a wide range of temperatures and pressures. The absorption coefficient of mixture was found to vary significantly over change of mixture temperature and probe laser wavelengths. The absorption coefficients of air/CNG mixtures were measured using 20 watts CW/pulsed CO₂ laser at 10.6μm, Pulsed Nd:Yag laser at 1.06μm, 532 nm (2^nd harmonic) and 4 mW CW HeNe laser at 645 nm and 580 nm for temperatures varying from 290 to 1000K using optical transmission loss technique.

Modern production processes are directed by properties of the components to be manufactured. These components have different sizes, functionalities, high assembly complexity and high security requirements. The increasing requirements during the manufacturing of complex products like cars and aircrafts demand new solutions for the quality assurance - especially for the production at different places. The main focus is to find a measurement strategy that is cost effective, flexible and adaptive. That means a clear definition of the measurement problem, the measurement with adapted resolution, the data preparation and evaluation and support during measurement and utilisation of the results directly in the production. In this paper we describe flexible measurement devices on example of three different techniques: fringe projection, fringe reflection and shearography. These techniques allow the detection of surface and subsurface defects like bumps, dents and delaminations with high resolution. The defects can be optically mapped onto the object's surface. Results are demonstrated with big components taken from automotive and aircraft production. We will point out the most important adaptations of the systems to realize miniaturized, robust and mobile devices for the quality assurance in an industrial environment. Additionally the implementation into a Mobile Maintenance and Control structure is demonstrated.

The selection of rover size, whether the environment be on land, in the sea or air, or on the surface of another world, necessarily entails certain tradeoffs. These tradeoffs include vehicle mass, power source, speed, range, size of obstacles that can be dealt with, sensor compliment, and ultimately, mission objectives. Smaller sized vehicles have advantages in that they tend to be more nimble and can more closely explore a complex environment, but, in general, at a cost of reduced capability in all other areas. Larger vehicles enhance these capabilities, but at a cost of being somewhat ponderous, especially in complex environments. Hierarchical roving seeks to maximize the best of these extremes by carrying a hierarchy of smaller specialized rovers within a larger one. The larger rover acts as a carrier vehicle, communications relay, and power recharge source. The smaller specialized vehicles are deployed at a given site, execute their mission, are then recovered by the carrier vehicle, and finally transported to the next site. Greater situational awareness and the opportunity for self-rescue are additional benefits of hierarchical roving. Experience with a carrier vehicle containing three smaller vehicles is discussed, as are the design tradeoffs.

This paper describes an innovative, high throughput manufacturing test system for testing high power laser-diode stacks. These stacks are based on a single high power bar building block, which can be stacked either vertically or horizontally to deliver extremely high output power (>3kW) from a compact package which can range from a single bar to over 25 bars in one package. Testing these various form-factors presents many challenges in high-volume manufacturing e.g. repeated changes of tooling and set-up to accommodate mixture of configurations. The automated test system described in this paper can accommodate any configuration of multi-bar stacks to test critical optical characteristics (LIV, Optical Spectrum Characteristics, Optical Power, Optical Divergence, water flow rate, water pressure etc.). Key to the automated station is a custom designed integrating sphere and universal stack holder with automated water flow configuration. The automated test system significantly improves the throughput by decreasing the test time by 50% (compared to manual testing). Individual bars comprising stack have different spectrum and the custom designed integrating sphere enables accurate spectrum analysis (centroid wavelength, FWHM) of the combined spectrum, as well as accurate power measurement.

Customer feedback in the form of warranty/field performance is an important and direct indicator of quality and robustness of a product. Linking warranty information to manufacturing measurements can identify key design parameters and process variables (DPs and PVs) that are related to warranty failures. Warranty data has been traditionally used in reliability studies to determine failure distributions and warranty cost. This paper proposes a novel Fault Region Localization (FRL) methodology to map warranty failures to manufacturing measurements (hence to DPs/PVs) to diagnose warranty failures and perform tolerance revaluation. The FRL methodology consists of two parts: 1. Identifying relations between warranty failures and DPs and PVs using the Generalized Rough Set (GRS) method. GRS is a supervised learning technique to identify specific DPs and PVs related to the given warranty failures and then determining the corresponding Warranty Fault Regions (WFR), Normal Region (NR) and Boundary region (BND). GRS expands traditional Rough Set method by allowing inclusion of noise and uncertainty of warranty data classes. 2. Revaluating the original tolerances of DPs/PVs based on the WFR and BND region identified. The FRL methodology is illustrated using case studies based on two warranty failures from the electronics industry.

This paper outlines the need for development of human posture prediction models for Computer Aided Design (CAD) and Computer Aided Manufacturing (CAM) design applications in product, facility and work design. Challenges facing developers of posture prediction algorithms are presented and discussed.

While injuries from powered hand tools are caused by a number of factors, this study looks specifically at the impact of the tools design features on perceived tool usability and safety. The tools used in this study are circular saws, power drills and power nailers. Sixty-nine males and thirty-two females completed an anonymous web-based questionnaire that provided orthogonal view photographs of the various tools. Subjects or raters provided: 1) description of the respondents or raters, 2) description of the responses from the raters, and 3) analysis of the interrelationships among respondent ratings of tool safety and usability, physical metrics of the tool, and rater demographic information. The results of the study found that safety and usability were dependent materially upon rater history of use and experience, but not upon training in safety and usability, or quality of design features of the tools (e.g., grip diameters, trigger design, guards, etc.). Thus, positive and negative transfer of prior experience with use of powered hand tools is far more important than any expectancy that may be driven by prior safety and usability training, or from the visual cues that are provided by the engineering design of the tool.

Fiber reinforced polymers are widely used in the transportation, aerospace and chemical industries. In rare instances these materials are produced net-shape, and secondary processing such as machining and assembly may be required to produce a finished product. Because fiber reinforced polymers are heterogeneous materials, they do not machine in a similar way to metals. Thus, the theory of metal machining is not valid for the analysis of machining of fiber- reinforced composites. Previous attempts in modeling this problem have adopted Merchant's theory from metal cutting by assuming that chip formation takes place in a shear plane which inclination angle is determined by the minimum energy principle. This class of models showed that model predictions are valid only for fiber orientations less than 60°. The work presented here focuses on providing predictive models for the cutting forces in unidirectional composites. The models are based on the specific cutting energy principle and account for a wide range of fiber orientations and chip thickness. Results from two forms of non-linear modeling methods, non-linear regression and committee neural networks, were compared. It was found that committee neural networks provide better prediction capability by smoothing and capturing the inherent non-linearity in the data. The model predictions were found to be in good agreement with experimental results over the entire range of fiber orientations from 0 to 180°.

A case study revealed that more than 13,500 MMBtu of energy is wasted annually when a single galvanizing line is off-production for hardware replacement for duration of a few hours every 2 weeks. This energy if utilized for production will yield about 13,000 tons of Galvanized Sheet Steel annually from a single galvanizing line. Thus for the 57 [1] hot dip galvanizing lines in US this figure results in a production loss of 741,000 tons/year. An attempt has been made to develop a spreadsheet that will take into account all the major energy consuming equipment in a typical hot dip continuous line. It maintains a track of the current production and energy consumption. It can simulate a scenario where either the number of shutdowns or the hours per shutdown will be reduced as a consequence of better material developed by the researchers. Different charts pertaining to energy consumed by different equipment group, total cost of energy spent on natural gas and electricity, MMBtu/Ton, Tons/Year and Production time before shutdowns assists the engineers decide the best operating stretch to suite their production rate and optimize energy consumption to some extent. Validation data gathered from the three well established galvanizing lines powers this spreadsheet to forecast annual increase in production and thus helps judge the performance of the new hardware.

The structural synthesis of planar mechanisms with dyads is presented. A classification of dyads is elaborated. New planar mechanisms are constructed using this classification of dyads. The trajectories for different mechanisms are studied. The advantage lies in the classification of the kinematic chains. The solution of the whole system can then be obtained by composing partial solutions.

Color and texture has been extensively studied in the field of image processing and computer vision. Industrial applications based on computer vision that uses color and texture information for produce recognition and surface change detection are more common these days. Even though color and texture have been individually studied and used for retrieval and classification purposes, very little work has been done in the problem of effective integration of color and texture information. Previously, we proposed the Collective Color Texture (CCT) model that functionally considers both the color and texture outcomes and generates an effective descriptor for the color texture. We showed that the CCT model outperformed other common integration methods when used for supervised classification. In this work, we use the CCT model for texture retrieval using histogram based color representation, various texture based representations. We used Outex 13 database for our experiments since it has wide variety of color textures (such as granite, canvas, carpet, etc) that are commonly present in industrial applications. We compare retrieval performance using individual methods with those from commonly used integrated techniques. Our results show that the CCT model provides an overall superior retrieval performance when compared with other popular approaches.

The performance of data and telecommunication equipment must keep up with the increasing network speed. Optical interconnection technology is a promising alternative for high throughput systems. The Optical backplane system was demonstrated with waveguide-embedded optical backplane, transmitter board and receiver board. The transmitter and receiver module were prepared for optical PCB, which consists of the metal optical bench, the driver chips, VCSELs, photodiodes and a tapered polymeric waveguide. And parallel optical transmitter and a receiver module were attached onto the processing boards for the interconnection with optical backplane board. The tapered polymeric waveguides are fabricated using the hot embossing technique. And the propagation loss of the waveguide was approximately 0.1dB/cm at 850nm. The waveguide-embedded optical backplane boards were fabricated by using conventional PCB lamination process. The data transmission characteristics of the processing board have been investigated. In our optical backplane system, we demonstrated up to 10Gb/s 2⁷-1 PRBS NRZ data transmission from the transmitter board to the receiver board through optical backplane. The BERs were less than 10^-12 under 8Gb/s data rate, which is sufficient level for telecommunications.

The rotary photoelectric encoder can be used in many fields, such as robot research, fruit assembly lines, and so on. If there have many photoelectric encoders in one system, it's difficult to manage them and acquire the right pulse number. So it's important to design a multifunctional management system. It includes a powerful microchip with high processing speed, assuring the acquisition precision of rotary pulse. It uses a special method to judge the rotary direction and will be competent for many occasions which rotary direction changes quickly. Considering encoder data transmission, the management system provides a serial port using RS-485 protocol to transmit current pulse data and rotary direction. It allows linking a maximum of 100 management systems using only two communication lines to up-systems and also configing the encoder counting pattern locally (using the keyboard) or remotely (through the computer).

The paper focuses on the current issues in the areas of computer aided manufacturing and design. The importance of integrating CAD and CAM is analyzed. The associated issues with the integration and recent advancements in this field have been documented. The development of methods for enhancing productivity is explored. A research experiment was conducted in the laboratories of West Virginia University with an objective to portray effects of various machining parameters on production. Graphical results and their interpretations are supplied to better realize the main purpose of the experimentation.

The objective of this research is to develop an intelligent information system to recognize rice seeds variety based on image processing and artificial neural network (ANN). At first, images of rice seeds were acquired with a color machine vision system. Each image was processed to extract twenty-one quantitative features. The classification ability of all the features was evaluated for different varieties recognition. To ensure system reasoning veracity and intelligence, ANN was used. A digital image-processing algorithm was developed to classify varieties of rice seeds based on external features. As an example of application of the system, the image data of common rice seeds varieties in Zhejiang province were input and the results of practical application proved that this software system achieved desired results.