Duplication is a process of making a copy of a copy of an existing object. To obtain a drawing of this object, the profile of the object has to be determined, followed by manual drawing to produce the drawing. Thus, this thesis presents a method that can help to automate the process of part drawing with the help of a vision system. The approach is useful and the time required to do shape detection of an arbitrary profile of an object can be reduced. By using the approach of a vision assisted automated CAD drawing, the tedious task of this process can be done easily. In the developed prototype system, the object which its drawing is to be produced automatically, is placed under the camera frame of a vision acquisition system and the image of the object is acquired. Later, an image processing module performs the necessary image manipulation to obtain the information regarding the object in the acquired image. This information is used by the automatic drawing module to produce the drawing of the object automatically. Finally, the 2D drawing of the object is drawn automatically.

Information Not Available

Information Not Available

Autonomous vehicles will be seen in the very near future on our roadways. When integrated with other advanced technologies like Global Positioning System (GPS), data communication system and so forth, a system with certain level of intelligence can be developed. The vehicles can be maneuvered automatically with highly optimized journey in terms of money, time, quality and efficiency. Intelligent braking system is one of vehicle subsystems that can be unprompted to function when needed. The system must be able to sense certain parameters like the speed of target vehicle and its own speed. From these measured variables the output brake force can be generated. To do this operation the system ought to have a knowledge base so as to have decision-making ability. The problem of modeling brake system is difficult due to its nature that inherits high nonlinear dynamics. It is possible, although laborious to come with analytical model of the system after severe constraints being imposed on the modeling. Technique of soft computing allows natural languages to play a role in scientific theory. It marks a significant paradigm shift, especially on how people see their analytical problems. Fuzzy logic can be an excellent tool to control a complex, nonlinear system. In this research, this tool is used to govern the brake system. Another aspect that has been tackled by this research work is the use of digital signal processor to run the fuzzy algorithm. It has several advantages over universal microcontroller and a dedicated fuzzy chip. The on chip peripheral allows the system to be composed in compact circuitry suitable for real time embedded system. It also provides room for other signal processing algorithms to be run within the same processor without any burning of cash needed to buy other processors. The result shows that the TMS320LF2407 processor can be a good processing engine for the braking system. It accords the simulation results with good accuracy. The time of Fuzzy Logic Inference Per Second (FLIPS) is also less than the time of scanning the input data.

Information Not Available

Information Not Available

Information not available

Information Not Available

Information not available

Differentiated Services have been proposed as scalable solution that can satisfy real time traffic. In a diffserv domain an Edge Router perfonns Multi Field classification and maintain per now profile. If a mobile node connects to a Diflserv domain, it is subjected to the same policing and shaping as done for the static nodes. When a mobile node moves and changes its connectivity from one RER to another, the new R ER does not have the context unless it is transferred from the old RER. A study is conducted considering a diffserv domain where RERs run three different marking schemes (TSW, srTCM, trTCM) to meler the flows and mark the packets. This thesis present~ the results of this study. which show that by transferring context during handoff leads to improvement in marking the now especially for TSW where the marking of packets reach stability qu1ckly and result in less delay and jitter. If the context is not transferred. lhen the marking at the new RER takes a while to reach stability. The instabi lity period is proportional to the window size setting. The results also show that the instabi lity caused by the hand ofT negatively effect the flows of the same class at the new RER.

Information Not Available

Information Not Available

Current anti-virus technology is largely reactive and relying on finding a particular virus before being able to deal with it. As virus numbers increase, a reactive approach becomes ever more difficult. Most of the current antivirus products rely upon scanning. The scanning technique has several problems which include capacity problems, update requirements and polymorphic virus. Capacity problems where the number of viruses discovered is growing at an ever-increasing rate. Update requirement problem where in order for a virus to be detected by a signature scanner it must first have been discovered and included in scanner updates. And polymorphic virus problem which are viruses that modify themselves in such a way that identification using a signature becomes more difficult or even impossible. Currently, there are many different types of computer viruses; this research concentrates on the viruses that attack executable programs. A design for the prevention and detection of computer virus has been developed based upon an advanced integrity checking method. This is based upon the fact that a sufficiently advanced change-detection system, which considers all factors including system areas of the disk and the computer memory into account, has the best chance of detecting all current and future viral strains. The expected benefits of this approach are: to prevent most of the current and future viruses from infecting a computer; to integrate with the operating system to provide an extension to the operating system so as to enforce security to the operating system’s resources and to be able to detect viruses that has no known signature.

Information not available

A study of shape control of piezoelectric laminated composite plate has been conducted in this thesis by employing the finite element analysis. Elastic field and electric field of the piezoelectric laminated composite plate has been coupled through the linear piezoelectric constitutive equations. Piezoelectric actuators and sensors are modeled as additional layers either to be surface bonded or emeded in the laminated composite plate. A higher-order shear displacement theory (HSDT) developed by Pervez (1991) has been employed in order to accurately describe the kinematic behavior of the piezoelectridc laminated composite plates. In the finite element model, an eight-noded two-dimensional isoparametric element is used. Each node has seven mechanical degrees of freedom and one electgrical degree of freedom. A computer code was written in C++ based on the finite elelment model, and was successfully and comprehensively validated with experimental and numerical results that are readily available in the literatures. The study of shape control has been done in two parts: open-loop shape control and closed-loop shape control. The effect of boundary condition, actuator voltage, fiber orientation, length to thickness ratio of actuator, segmented actuator and actuator orientation have been analyzed for the open-loop shape conrol. Under closed-loop shape control, the present research demonstrates the ability of the piezoelectric composite laminated plate to actively control its shape under varying loads.

Information Not Available

Information Not Available

One of the most important capabilities of MPLS is to map traffic on paths established with traffic engineering principles, resulting in load balancing and fault tolerance of the underlying network. Traffic engineering deals with performance optimization and evaluation so that the overall performance for a network can be enhanced. Traffic Engineering with Link Coloring (TELIC) is an algorithm that assigns LSP’s to incoming traffic requests that enter an MPLS-DiffServ domain in a service aware manner. TELIC routes the best effort traffic preferably on separate links, away from the links carrying premium traffic. TELIC is shown to achieve remarkable success in routing EF or premium trunks separately with the DF flows. However, the DF rejection ratio is higher in TELIC as compared with other TE algorithms. This thesis proposes an enhancement for TELIC and SHORTD in order to increase the bandwidth allocation ratio for best effort traffic. Default Forwarding Trunk Splitting (DFTS) is a new algorithm developed and integrated with TELIC and SHORTD. This algorithm splits the rejected DF trunks across the domain into several smaller trunks and routes them on separate preferred links. DFTS has been developed in Visual C++ using object oriented design. TELIC’s performance with and without DFTS algorithm is measured by using it on a specific set of traffic requests that arrive at networks built with two different topologies such as ISP and Disjoint Multipath. The same traffic sets and topologies are used to measure SHORTD’s performance with and without DFTS.

A new effective method for the microchip lead inspection for the chip manufacturing industry has been developed in this research. This new technique is implemented using Lab View programming environment and tested on many sample images using National Instruments frame grabbers and hardware interfaces. The first part of the software deals with the image enhancement and auto-thresholding method to prepare for precision measurements. The second part of the software is using newly developed statistical analysis technique for automatic IC leads defect detection (such as linearity, implied planarity, offset...etc). Finally the research carries out a precision, repeatability and reproducibility test procedures of the system. Contrary to the gray scale pattern matching technique, the proposed technique employs selected parameters of binary blobs to perform fault detection and measurements. This leads to a significant reduction of image processing time. A special combination of gray level filtering techniques with gray morphological operations has been used to enhance the borders of the lead images. A newly developed threshold calibration technique significantly improves the measurement accuracy. A unique statistical analysis has been developed to identify all possible lead defects in the chips. This method is rotationally and scale invariant and able to detect defective leads for the chips with different specifications. The minimum required information about the microchip is the number of leads.

Information Not Available