Purpose -The paper aims to propose a system that uses a combination of techniques to suggest weld requirements for ships parts. These suggestions are evaluated, decisions are made and then weld parameters are sent to a program generator. Design/methodology/approach -A pattern recognition system recognizes shipbuilding parts using shape contour information. Fourier-descriptors provide information and neural networks make decisions about shapes. Findings -The system has distinguished between various parts and programs have been generated so that the methods have proved to be valid approaches. Practical implications -The new system used a rudimentary curvature metric that measured Euclidean distance between two points in a window but the improved accuracy and ease of implementation can benefit other applications concerning curve approximation, node tracing, and image processing, but especially in identifying images of manufactured parts with distinct corners. Originality/value -A new proposed system has been presented that uses image processing techniques in combination with a computer-aided design model to provide information to a multi-intelligent decision module. This module will use different criteria to determine a best weld path. Once the weld path has been determined then the program generator and post-processor can be used to send a compatible program to the robot controller. The progress so far is described.

Improvements are described for a pattern recognition system for recognizing shipbuilding parts. This is achieved by using a new simple and accurate corner finder. The new system initially finds corners in an edge detected image of a ship's part and uses that new information to extract Fourier descriptors to feed into a neural network to make decisions about shapes. Results show that the new corner finder was better at distinguishing between various ships' parts than other corner finders and proved to be a valid approach. The new corner finding technique uses a bottom-up approach to find corners by sampling points in edgedetected images and calculating the distance between the endpoints of a window around each sampled point. The points with the minimum distance are then interpreted as corners. Using an all-or-nothing accuracy measure, the new corner finding technique achieved an improvement over other systems. The new corner finder was included as pre-processing before extracting Fourier descriptors and using the artificial neural networks to identify parts. The whole system recognized parts more quickly and more efficiently than the most recently published systems.

No abstract

No abstract

To achieve the flexibility required by the small amount of repeated work in some industries, automated systems should be capable of being reprogrammed quickly and efficiently. One way of doing this is to create software dedicated to the task of generating the required code. Using Object-Oriented Programming techniques, a software engineer can write efficient code for machines that is faster to implement and extendable. This paper gives a brief overview of objectoriented programming and then goes on to discuss research into the use of that technique to create programs to generate code for a Motoman welding robot in the shipbuilding industry.1.