Optical section
The lens assembly is a complex ensemble of precision glassware, with adjustment for focus and zoom by sliding members within the barrel. Both focus and zoom rings are driven by miniature electric motors; where manual (hands-on lens ring) adjustment is provided for these, the motors drive through slipping clutches. Zoom motor control is provided for in a body-mounted rocker switch, sometimes offering two zoom speeds. The auto-focus system consists of a servo loop with through-the-lens (TTL) picture-sharpness sensors. It is controlled by a microprocessor, and many models offer a choice of picture zones for auto-focus operation.
Also inside the lens assembly is a multi-bladed iris, whose operation is similar to that of a moving-coil meter. It too is part of a servo loop, this time controlled by the level of luminance signal coming from the image sensor. The higher the light level in the televised scene the smaller the iris opening, with a consequential improvement in depth of focus field. There is also an AGC system in the video amplifier: between these two control loops the signal level—and hence contrast in the reproduced picture—is held constant over a huge range of ambient scene brightness. In very low light situations, however, noise (grain, snow, confetti) intrudes on the picture.
Interposed between the lens and the image sensor are two optical fil- ters. An infra-red cut filter prevents most infra-red radiation reaching the sensor so that heat and similar energy sources have little effect on the pic- ture, while a crystal filter takes out the finest detail in the incoming scene to prevent ‘beat’ and patterning effects due to the dot-matrix structure of the sensor IC and its colour filter and to minimise cross-colour effects.
CCD image sensor
The CCD image sensor consists of a mosaic of photodiodes, each of which acquires and stores a charge proportional to the intensity of light falling upon it. The charges are transferred by horizontal and vertical CCDs to the image-sensor’s output terminal, whence they emerge (after processing in a sample-and-hold stage) as a serial video signal. The colour components of the image are captured by a coloured translucent dot-matrix overlay bonded to the face of the pick-up sensor chip. Digital colour cameras gen- erally use a Bayer mask over the CCD. Each square of four pixels has one filtered red, one blue and two green (the human eye is more sensitive to green than either red or blue). The result of this is that luminance infor- mation is collected at every pixel, but the colour resolution is lower than the luminance resolution.
Better colour separation can be reached by three-CCD devices and a dichroic beam splitter prism, that splits the image into red, green and blue components as was explained in Chapter 12. Each of the three CCDs is arranged to respond to a particular colour. Some semi-professional digital video camcorders (and all professionals) use this technique.
Image stabilisation
A problem with very small camcorders is shake and wobble in the repro- duced image due to the natural tremor in the human hand which holds it. It is especially troublesome at extreme zoom settings. To mitigate this,
494 Newnes Guide to Television and Video Technology
optical image stabilisation (OIS) is used. There are different techniques that have been developed. One brought to the market by Sony, involves a vari- angle refractive prism, consisting of two silicon–oil filled plate-glass pan- els linked by bellows, mounted in the lens assembly. The prism’s angle, relative to the pick-up sensor’s image plane, is varied by drive coils under the control of a piezo-electric sensor of pitch and yaw of the camcorder’s body. Another technique use DSPs to analyse the image on the fly and then move the sensor appropriately.
A third electronic technique depends on the use of an A-D converter in conjunction with a large DRAM digital memory in which one whole field of picture information is temporarily stored. The image for recording is read from the centre section of the memory bank, using approximately 85% of its contents, area-wise. The pixel data corresponding to the outer periphery of the picture is selectively used by the electronic image stabiliser (EIS) processor with reference to four motion-detection zones in the pic- ture and an algorithm which distinguishes between camera shake and nat- ural movement in the picture.
Digital image stabilisation is used in some video cameras. This tech- nique shifts the electronic image from frame to frame of video, enough to counteract the motion. It uses pixels outside the border of the visible frame to provide a buffer for the motion. When the camera moves, the whole frame changes position on the chip, but it still registers in its entirety. The camera’s computer picks a fixed point in the picture, like the outside edge of the frame, then stabilises that edge as a fixed point on the recording.