Anatomy of the MIC-D Digital Microscope
Electrical and Image Sensor Control System
Because the Olympus MIC-D digital microscope operates with the assistance of electrical current derived from a host computer, it does not require batteries or an external alternating or direct current power supply. The microscope has a rated operating potential of 5.0 volts, a nominal current draw of 0.4 amperes, and receives power through a Universal Serial Bus (USB) interface from the local host computer. The power supplied by the host computer serves to provide current to the microscope illumination and image sensor components, which are the primary electrical (and electronic) systems that control luminous flux and image capture with the microscope.
Components in MIC-D microscope electrical system include the light-emitting diode (LED) microscope illuminator, the associated lamp voltage control circuit board, and an image sensor control circuit board (all are illustrated in Figure 1). Power to these circuit boards is controlled through a switch integrated into a potentiometer positioned in the base of the microscope. In addition, a second potentiometer is geared to the zoom handle to increase or decrease lamp voltage as the microscope magnification is changed. Power for the CMOS image sensor is also derived through the USB port, which is integrated into the circuit board containing the sensor photodiode array integrated circuit.
As presented in Figure 1, twin shielded electrical wires are routed from a connector on the illumination control circuit board through the microscope base into the zoom handle, where they are attached to one of the supporting rods with a section of heat-shrink tubing. These wires then pass through the upper microscope body and into the rotation arm, where they are terminated by a female connector in the center of the arm. A second set of wires originating from the LED lamp, also attached to a female connector, meets the first connector and is joined by a double male connector inside the rotation arm shroud. In order to access these connectors (to change lamps), the four Phillips-head screws that secure the polymer shroud to the rotation arm must be removed. The loose shroud can then be gently lifted from its seat to access the connectors.
The MIC-D digital microscope has a visible light source consisting of a gallium-nitride based white light-emitting diode contained within the illumination head, as illustrated in Figure 2. Electrical power for the LED is derived from the twin wire cables that snake through the body of the microscope as previously described. Access to the diode lamp is gained through a molded polymer access cover (see Figure 2), which is threaded to fasten on top of the illumination head. The lamp is mounted in the center of a 5-millimeter aperture positioned in a circular aluminum bracket within the illumination head, and secured by two Phillips-head screws having polymer washers.
The light-emitting diode illumination source produces white light through the interaction of a cerium-doped yttrium aluminum garnet (YAG) phosphor material with blue light emitted by the semiconductor diode. The result is a broad spectrum of emission in the visible region with a spike (maximum) in the blue wavelength region at approximately 460-470 nanometers. These lamps typically have a brightness of 5 lumens per watt and provide adequate illumination for specimen observation with the MIC-D digital microscope in all imaging contrast modes. The average lamp lifetime can exceed 100,000 hours, which dramatically reduces the frequency of replacement. In fact, many of the diodes may have a lifetime that exceeds that of the other electrical components in the microscope.
As discussed above, electrical power for the microscope is derived from a host computer through a Universal Serial Bus interface, which is a commonly utilized industry-standard specification for attaching peripheral components to a computer system. The USB interface delivers high performance and enables the ability to connect devices, such as the MIC-D microscope, while the host computer is operating. A 2-meter USB cable supplied with the microscope has non-symmetrical connectors on each end, which are illustrated in Figure 3. The cable connects to the host computer through a type A male USB connector, while a type B USB connector attaches to a port in the image sensor control board (Figure 5) housed in the microscope base.
Two ferrite cores are provided with the MIC-D digital microscope to prevent interference of the communications and power signal between the microscope and host computer by extraneous electromagnetic radiation. The cores are both placed near the type B USB connector at a distance of 2 to 4 centimeters apart. One core is designed to attach to the cable adjacent to the rear of the connector in a sandwich over the wire, while the other should have the cable wrapped around the periphery.
Windows 2000 software drivers for the USB interface are provided with the MIC-D digital microscope, although the latest versions are also readily available from the operating system manufacturer. The USB specification provides for a maximum rate of 12-megabit per second data transfer, and supports up to 127 external devices that have hot-plug and hot-unplug capabilities. In addition to the MIC-D microscope, other devices simultaneously utilizing the USB interface can include color inkjet printers, laser printers, flatbed and film scanners, digital still cameras, video cameras, modems, keyboards, mice, joysticks, game pads, hard disk drives, removable storage devices, digital speakers, and digital microphones.
The MIC-D digital microscope CMOS image sensor is installed on a circuit board (the Image Sensor Control Circuit Board; see Figures 4 and 5) and is aligned with the optical axis of the microscope. Light focused through the microscope optical system is captured by the CMOS photodiode array to generate electrical charge that is interpreted into an image by the controlling electronics. The circuit board containing the image sensor integrated circuit also contains the USB interface connector and supporting electronics.
At the heart of the MIC-D image sensor control circuit board is an OmniVision OV511+ Advanced Camera-to-USB (ACUB) single-chip controller integrated circuit (see Figure 5) designed for video applications using the USB interface. This chip dramatically simplifies the interface between single-chip CMOS image sensors and the bus system, enabling the design of a complete USB-based video subsystem. An on-chip proprietary high-performance compression engine capable of a 7:1 image compression ratio ensures that fast image transfers can occur between the sensor and the host computer. In addition, the controller also supports high-speed decompression in the host system.
The OV511+ provides two ports that accept 16-bit YUV 4:2:2/RGB raw data, and a third port that interfaces to devices that output 8-bit YUV 4:0:0/RGB raw data (not utilized by the MIC-D digital microscope). Control lines, including vertical sync, horizontal reference output, and clock inputs are provided for integrated CMOS camera chip management. The interface circuit is supported by 4 megabytes of video memory and complete USB device controller and system controller functions. On-chip registers are programmed over the OV511+ ACUB parallel input/output bus. The CMOS image sensor and OV511+ co-processor perform the key functions of image capture, digital video image processing, video compression, and interfacing to the host computer system via the USB port.
Image capture in the MIC-D digital microscope is orchestrated by an OmniVision CMOS image sensor that combines all necessary camera functions onto a single integrated circuit. Automatic controls include automatic gain control (AGC), automatic exposure control (AEC), automatic white balance (AWB), gamma correction, automatic level control (ALC) and automatic black level calibration. In addition, the image sensor includes on-chip color interpolation into the YUV 4:2:2 color space. The image sensor has a photodiode array with 640 x 480 sensor elements (307,000 active photodiodes) in a standard one-third inch format. An important feature of the CMOS sensor is the sub-sampled image size of 320 x 240 pixels (77,000 pixels) that enables smaller images to be captured with the microscope.
The MIC-D digital microscope CMOS image sensor is designed with square pixels, and is capable of both progressive and interlaced scanning modes (user selectable). There are two on-chip 10-bit analog-to-digital (A/D) converters that provide YUV or raw RGB data in digital output data streams of 8-bit or 16-bit widths (the MIC-D utilizes the latter). Early versions of the microscope utilized an image sensor with 8.4-square micron pixels, while more recent versions have a more advanced chip having pixel dimensions of 7.6-square microns. The new chips increase the microscope magnification by a factor of 11 percent.
The camera system is initialized and controlled through an EEPROM programmable read-only memory chip that also receives information from the computer through the USB port. Timing functions for the circuitry are performed on the image sensor with the assistance of several crystal oscillators placed as surface mount components on both the camera and illumination control circuit boards. The OV511+ coprocessor retrieves image data from the CMOS image sensor and then processes the data before passing it along to the USB port. The coprocessor also features hardware video compression that is capable of producing video frame rates 30 frames per second at the lower (320 x 240) resolution, and 10 to 15 frames per second at VGA (640 x 480) resolution.
The OV511+ coprocessor supports isochronous data transfer mode, which is an implement of the USB specification. Isochronous (derived from the Greek words for equal and time) processes refer to data transmissions that require exact timing coordination to be successful, such as multiplexing audio with digital video information. This transfer mode is useful in multimedia streams that demand data be delivered at the same rate as the computer display paint to ensure that audio is synchronized with the video. Both the high-performance USB and the asynchronous transfer mode (ATM) specifications support isochronous data transfer.
Positioned adjacent to the image sensor control circuit board is the light-emitting diode illumination control circuit board (see Figures 4 and 6). This small board contains components that regulate the luminosity of the white LED, and thus the quantity of light passed through the optical system of the microscope. Adjustment of LED intensity (luminosity) is controlled by two potentiometers. The main power switch for the microscope is attached to a potentiometer that has a control knob extruding from the microscope base. This variable resistor controls the voltage to the LED. A second potentiometer is geared to the zoom handle and is designed to adjust LED voltage as the magnification is increased to provide even illumination throughout the zoom range. Adjustment of luminosity by the electronic circuitry is performed with pulsed lighting in order to avoid influencing the color balance of the image.
An advantage of utilizing CMOS optical image sensors in peripheral devices, such as the MIC-D digital microscope, is their low power consumption requirements. Chips on the circuit board and the light-emitting diode illumination system of the MIC-D require so little energy that the microscope can obtain power directly from the USB input/output interface. Image sensor technology surrounding the CMOS designs is becoming a fiercely competitive business with big name players such as Intel, Lucent, Kodak, and Hewlett-Packard jockeying for position.
The CMOS active pixel sensor, and related designs, is an emerging technology that will push the limits on digital imaging both in toys and in the general consumer electronics market. High-performance models are already making inroads to the professional camera level, and may soon be competing with charge-coupled device (CCD) technology. Mass production of CMOS devices is relatively economical and many facilities that are currently engaged in fabrication of microprocessors, memory, and support chips can be easily converted to produce CMOS optical image sensors. Although CCD chips were responsible for the rapid development of video camcorders, the technology has remained trapped as a specialized process that requires custom tooling outside the mainstream of integrated circuit fabrication. Also, the CCD devices require a substantial amount of support circuitry, and it is not unusual to find five or six circuit boards in a typical CCD camera system.
The primary concerns with CMOS technology are the rather low quality, noisy images that are obtained with respect to similar, competing CCD devices. The artifacts are due primarily to process variations that produce a slightly different response in each pixel, which appears in the image as snow or fixed pattern noise. Another problem is the total amount of chip area that is sensitive to light is less in CMOS devices, making them far less sensitive in fixed illumination conditions. These problems will be overcome as process technology advances and it is very possible that CMOS devices will eclipse the CCD as the technology of choice in the very near future.
Thomas J. Fellers and Michael W. Davidson - National High Magnetic Field Laboratory, 1800 East Paul Dirac Dr., The Florida State University, Tallahassee, Florida, 32310.
Questions or comments? Send us an email.
© 1995-2018 by Michael W. Davidson and The Florida State University. All Rights Reserved. No images, graphics, software, scripts, or applets may be reproduced or used in any manner without permission from the copyright holders. Use of this website means you agree to all of the Legal Terms and Conditions set forth by the owners.
This website is maintained by our