System integration extends imaging capabilities
System integration extends imaging capabilities
George Kotelly, Executive Editor
When several machine-vision and imaging products are interfaced and integrated into complete vision systems they are capable of solving a variety of detection, diagnostic, identification, and inspection problems. Using a noninvasive, nuclear medical-imaging technique known as coincidence imaging, positron emission tomography can distinguish cancerous from noncancerous tissues as well as determine the growth rate of cancerous lesions. Contributing editor Kathy Kincade reports that integrated data-acquisition, imaging, curved detector plate, and display hardware can now create a 3-D image that clearly indicates the physiology, location, and distribution of the region of interest (see p. 35).
To screen airline travelers for lawful entry, the United States Immigration and Naturalization Service, with support from other government agencies, has developed and installed automated imaging inspection systems at US airport terminals. Says contributing editor R. Winn Hardin, these systems validate biographic and biometric data and digital images against a computer-stored database to correctly identify several-hundred-thousand travelers and tourists (see p. 47).
In semiconductor fabrication, wafer alignment and identification generally involve two steps because of the size of the equipment needed. To solve this problem, three system integrators have interfaced their hardware and software to perform both steps with one machine-vision platform. According to contributing editor Larry Curran, this integrated system captures and validates an image of the identification marks on the wafer`s surface (see p. 57).
In many image-processing applications, feature- and pattern-recognition techniques are used to match specific features in an image with known templates. Some of these applications, however, require the simultaneous analysis of multiple, complex, and irregular features within images. By incorporating neural-network techniques, says editor at large Andy Wilson, developers can train imaging systems with numerous images until the system eventually "learns" to recognize irregularities (see p. 63).
The CompactPCI specification, a superset of the Peripheral Component Interconnect standard, uses the Eurocard form factor popularized by the VME bus. Says Andy Wilson in this month`s Product Focus, system developers are turning to CompactPCI frame grabbers to obtain rugged 3U and 6U form factors for harsh environments (see p. 73).