By Topic

Application of Machine Vision, IEE Colloquium on

Date 24 May 1995

Filter Results

Displaying Results 1 - 12 of 12
  • Application of machine vision in image compression

    Publication Year: 1995 , Page(s): 5/1 - 5/6
    Cited by:  Patents (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (320 KB)  

    A general-scene object-oriented image coder is presented which can encode colour TV signals at low bit-rate independent of content, complexity, and temporal behaviour of the TV sequence. This is a novelty compared to object-oriented coders developed by other researchers which are specialised in video-phone scenes. For the realisation of the coder's generality, stereoscopic and motion information has been used for segmentation and ordering of the objects. This coder is a new approach to the concept of object-oriented coding because it can encode without the use of pre-defined models and without a priori knowledge of the scene content. It is also unique in its type because it utilises machine vision knowledge for object segmentation and manipulation. This work was done within the framework of RACE (Research on Advanced Communications in Europe) under contract R2045 “DISTIMA” (DIgital STereoscopic IMaging and Applications) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Uncalibrated stereo vision for PCB drilling

    Publication Year: 1995 , Page(s): 4/1 - 4/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (344 KB)  

    Machine vision systems are often used in conjunction with industrial robots to automate manufacturing operations. However, such automation is limited by the need for accurate calibration of both the robot and the vision system. This paper discusses how the mathematics of projective geometry may be used in conjunction with a model of the workpiece to avoid the need for camera calibration. This in turn eases the integration of machine vision with certain industrial processes. Particular consideration is given to the application of this technique to printed circuit board (PCB) drilling View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 3-D vision systems using rotating 1-D sensors

    Publication Year: 1995 , Page(s): 6/1 - 6/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (252 KB)  

    This paper describes continuing research with both two-dimensional (2D) and three-dimensional (3D) rotating line-scan vision systems in industrial inspection and control automation. Traditionally, two-dimensional line-scan systems have been effectively used in a production line environment where movement is inherent in the application. However, when only using a single sensor, depth or range information cannot be ascertained from points of interest in object space. The research outlined in this paper describes how omni-directional range data from a scene of interest may be obtained when utilising a rotary 3D arrangement of line-scan sensors. The ability to resolve range information allows the location of any point to be determined within the field of view surrounding the camera arrangement. An advantage of a rotating line-scan vision system is that the field of view (and thus the resolution) of an image can be varied from any arc up to 360°. If an “all-round” field of view is set up, the images produced will contain information from an area of the object space completely surrounding the camera arrangement. This field of view may then be adjusted to cover a smaller arc and used interactively with the image start point. The resultant images can then display information at a given resolution from any part of the observed object workspace surrounding the rotational configuration. An obvious benefit of such a versatile system is, for example, the simultaneous control of a number of robot arms located around the camera arrangement. This paper presents results pertaining to the 2D and 3D aspects of the rotating line-scan research, including the potential accuracy of the system over a variety of stereoscopic configurations. This demonstrates the use of the line-scan system in specific applications where conventional sensor modalities may otherwise not be appropriate View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The automated inspection of moving webs using machine vision

    Publication Year: 1995 , Page(s): 3/1 - 3/8
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (284 KB)  

    Many materials are manufactured industrially in the form of a continuously moving strip or web; examples include metal strip (cold rolled, plated and stainless steel; aluminium), paper, glass, textile materials (woven, knitted and printed fabric; lace; carpet), etc etc. To ensure high product quality it is necessary to inspect visually,; this inspection can be automated using machine vision. The motivations include increased cost effectiveness, much higher operating speeds, reliability, consistency and objectivity, and improved record keeping. For some applications, automated inspection using machine vision is already operational; other cases remain to be conquered. This paper discusses progress in the application of machine vision to automate web material inspection, concentrating on the development of a generalised theory for signal processing for defect detection. Limitations to the present state-of-the-art in machine vision inspection are considered. It is suggested that the dominant factor limiting the spread of machine vision inspection in industry is cost effectiveness View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic casting surface defect recognition and classification

    Publication Year: 1995 , Page(s): 10/1 - 10/5
    Cited by:  Papers (2)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    High integrity castings require surfaces free from defects to reduce, if not eliminate, vulnerability to component failure from such as physical or thermal fatigue or corrosion attack. Previous studies have shown that defects on casting surfaces can be optically enhanced from the surrounding randomly textured surface by liquid penetrants, magnetic particle and other methods. However, very little has been reported on recognition and classification of the defects. The basic problem is one of shape recognition and classification, where the shape can vary in size and orientation as well as in actual shape generally within an envelope that classifies it as a particular defect. There have been many algorithm proposed for object recognition and classification based on such as neural networks, template matching, fuzzy logic, moment invariant methods etc. and all of these were tested by the authors for robustness and flexibility. From this initial study, the algorithms based on fuzzy logic emerged as having the most potential View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A computational model for learning to navigate in an unknown environment

    Publication Year: 1995 , Page(s): 9/1 - 9/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (364 KB)  

    The ultimate goal of this research is to design a system to automatically learn a visual domain so that it can subsequently execute a controlled navigation from any location to another when instructed to do so. The required system must have: flexibility, autonomy, scalability and robustness, which is defined for clarity. It must be flexible in order to cope with a broad range of problems, for example indoor and outdoor path planning and a large class of visual features i.e. the ability to function in as broad a range of circumstances as possible. We have based our algorithms on producing an architecture which can learn an unknown environment, using a self-generating map. A self-generating map is an extensible neural network method which uses self organising features. Our method is different in that is uses a novel method for feature extraction and a novel neural network architecture-the contextual layered associative memory View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IEE Colloquium `The Application of Machine Vision' (Digest No.1995/113)

    Publication Year: 1995
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (476 KB)  

    The following topics were dealt with: machine vision applications in industry; detection of contaminants in food products; automated inspection of moving webs; stereo vision for PCB drilling; machine vision in image compression; 3-D vision systems; arc weld join investigation; recovery of third dimension; navigating unknown environments; automatic casting surface defect recognition; and the application of parallel DSP networks to real time image processing View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Machine vision-applications in industry

    Publication Year: 1995 , Page(s): 1/1 - 1/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (272 KB)  

    Rapid advances in the automation of production methods have increased inspection requirements for three main reasons. Firstly, higher production speeds require higher inspection speeds. Secondly, the implicit inspection involved in manual production and assembly is no longer present and must be accommodated elsewhere and lastly, the ever-increasing demand of the customer for higher quality. Machines that can see have been developed for a variety of tasks that involve inspecting and manipulating industrial artifacts. The presentation discusses some of the problems faced by various industries and possible solutions using machine vision systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Sensitive X-ray detection of contaminants in food products

    Publication Year: 1995 , Page(s): 2/1 - 2/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (344 KB)  

    Over the past 20 years machine vision has grown from an infancy of exciting but mostly unrealisable prospects into a mature field in which sophisticated real-time systems can be installed economically in agricultural, industrial, medical, surveillance, transport and many other applications. In these areas it is commonly applied to active control, e.g. of guided vehicles and robots, but there are many other situations where it is used more passively to check for intruders, to keep a watch on the flow of people (e.g. on streets or on the underground) or to monitor industrial plant. In this last category falls the subject of automated visual inspection. Automated visual inspection is largely concerned with the maintenance of standards, and particularly to ensure that precision parts are made with the correct dimensions, the right number of holes in the right places. and so on. It is also concerned with checking that more complex parts have been assembled correctly and with the right tolerances, and in these respects it merges imperceptibly with automatic or robot assembly View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recovery of the third dimension using a priori geometric knowledge

    Publication Year: 1995 , Page(s): 8/1 - 8/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (220 KB)  

    The implementation of a real-time imaging system to support the robotic transfer of semiconductor wafers within an ion implanter is described. Since the system is to be embedded in a product, it must be economically viable, and therefore capable of being supported by an inexpensive computing platform. In practice, this can be achieved by exploiting the inherent constraints. In this application, a great deal of the geometry of the environment to be sensed is known a priori, which has resulted in the implementation of an optimal system configuration. This has been performed in a controlled manner, and thus the flexibility of the overall system has not been compromised View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The application of parallel DSP networks to real time image processing

    Publication Year: 1995 , Page(s): 11/1 - 11/6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (240 KB)  

    Image processing, and especially real time image processing, is naturally a very compute intensive task. These tasks can now be tackled very well with the Texas Instruments TMS320C40 DSP processor. This, as well as having a powerful processing architecture well suited to data intensive signal processing tasks such as image processing, also has six high speed communication ports for transferring data between C40 processors in a network. Coupled with suitable modular C40 based hardware such as that from Loughborough Sound Images, this can provide an extremely powerful and flexible image processing system. This paper presents some results of using C40 processors in practical image processing systems, both a single C40 and multiple C40s. It deals with the real processing speed achievable and techniques to achieve that speed, then the speed of transferring image data between C40s, for example partitioning a grabbed image across to several C40s. It then goes on to demonstrate how an application was developed making use of the above techniques, to give a real time image processing system processing a full 512×512 image at 12.5 frames per second View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Arc Weld Join Investigator

    Publication Year: 1995 , Page(s): 7/1 - 7/7
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (276 KB)  

    The Are Weld Join Investigator will scan the join between two pieces of piping to produce dimensional data that will aid the welding of the join. The idea for the project came from ISOTEK, a Leeds firm who lay and maintain oil pipes in the North sea. When, for example, a section of piping needs to be replaced a join must be welded to secure the new piping to the old. The welding, using an arc, must be done very accurately. Information is needed about the nature of the join to be made. The information required and the welding of the join cannot be done by divers at deep sea levels (for example 200m). Also there is a demand for very accurate welding (in the order of millimetres) which cannot be achieved manually. Only robots may be able to perform within these parameters. The project aims to use image processing techniques to find the exact geometry of the pipes. A dry run is performed whereby a video camera follows the join around the circumference of the pipe. A frame grabber collects complete images at a set rate (for example one per second) and makes the image data available for computer processing. The processing extracts dimensional information from the image data, and by using this information parameters may be set for the weld of the join (for example arc distance/angle from the pipe, arc current/heat intensity) View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.