Automated analysis enlightens imaging
Automated analysis enlightens imaging
George Kotelly Executive Editor
Until recently, many imaging-analysis tasks were performed manually and visually by skilled operators. Analysis studies involved the evaluation, measurement, and interpretation of numerous images and data. At high-information volumes, human errors and false results tended to increase.
Recently developed imaging inspection, analysis, and evaluation systems and software are automating these tasks and, at the same time, are also markedly improving the quality, consistency, and accuracy of test results. Moreover, automated image analysis translates into time and cost savings. Several such advanced machine-vision and image-processing systems are covered in this issue of Vision Systems Design.
To inspect clear molded or machine parts, liquid-crystal-display panels, lenses, and transparent components for surface and edge defects consistently, an automated inspection system combines cameras, LED lighting, image-processing boards, x-y positioning motors, controllers, and off-the-shelf illumination and imaging hardware. By additionally integrating a proprietary illumination /imaging subsystem and neural-network image-processing software, says contributing editor Larry Curran, the system achieves higher accuracies and speeds than human inspectors (see p. 22).
In pathology, image-analysis tools are attempting to automate tasks and thereby improve test time, quality, consistency, and results. According to contributing editor John Haystead, a special automated image-management system, under development for more than two years, is combining pathology images with patient information and imagery from radiology, microscopy, and ultrasound sources to provide precise digital data and to speed patient diagnosis (see p. 12).
Sophisticated medical imaging systems are automatically transforming large amounts of data from a variety of medical instruments into interactive visual displays of lifelike structural and functional human organs. As investigated by editor at large Andy Wilson, fast, powerful, and three-dimensional visualization software tools are helping physicians and researchers to better examine, explore, and detect organ defects (see p. 30).
Assisted by advances in DSP technology, image-processing boards are now providing higher levels of performance for automated machine-vision applications (see p. 40). These boards, says contributing editor Rick Nelson, are also accommodating one or more processors and delivering optimum price performance.