Like a CCD, a CMOS sensor is used as an image sensor in digital still and video cameras. It uses an internal array of photodiodes to convert images entering the camera through the lens into digital images. The major structural difference between a CMOS sensor and a CCD is the fact that each photodiode in a CMOS sensor has an amplifier that converts electrical charges to voltage. In a CCD, the photodiodes in each row act like a "bucket brigade." The charges passed along each row are collected and transferred to the output stage, where they are converted to voltage by an amplifier.

CMOS sensors have long been widely known as a type of image sensor offering a number of advantages over CCDs, including low power consumption, high data retrieval speed, and compatibility with LSI logic. However, for many years they were considered unsuitable for perusing high picture quality imaging applications because of issues relating to their structure and manufacturing processes.

Over the last few years there have been dramatic improvements in the image quality achievable with CMOS sensors, and today they are used in television cameras and professional SLR digital cameras. Like CCDs, CMOS sensors have started to attract increasing interest as image sensors. However, there are still major obstacles to the use of CMOS sensors in producing high resolution images.

Demand is on the rise for digital video cameras capable of capturing high resolution still images and HD video images. Naturally, such capabilities require high-resolution image sensors. At the time, the challenge was to increase resolution (by increasing the number of pixels) without increasing the size of the image sensor. At first glance, one would think that the only answer would be to reduce the size of each individual pixel. However, this reduces sensitivity because it results in a reduction in the light-receiving surface area of each pixel. Conversely, to increase sensitivity without changing the number of pixels (by increasing the size of each pixel), it would be necessary to increase the size of the image sensor itself, making it impossible to achieve highly-compact products. The goals of increasing resolution and improving sensitivity have generally been regarded as mutually exclusive, which meant that it was impossible to create a video camera, especially a compact video camera, capable of capturing high resolution still images. Sony overcame this challenge by developing and bringing to market its new ClearVid CMOS Sensor, and by optimizing its image processing engine, the Enhanced Imaging Processor, to maximize the performance of the new image sensor.

The most important feature of the ClearVid CMOS Sensor is its unique pixel array. Each pixel has been rotated 45º, and the color array has also been modified. By rotating each pixel 45º, Sony has almost doubled the size of each leading to enhanced sensitivity. The number of pixels appears to have been reduced, which would lead one to conclude that this arrangement cannot support advanced resolutions. However, the sensor has been designed to compensate by taking advantage knowledge regarding human eye perception, including the tendency of the eye to perceive images containing a high percentage of luminescence or "brightness" as having a higher resolution, and a higher sensitivity to vertical and horizontal resolution than to diagonal resolution. Green has the greatest influence on brightness, so Sony placed three times more green pixels (compared with other conventional sensors) around the other colors, thereby compensating for the loss of resolution.

In a camera with a single image sensor, not all pixels in the sensor carry red (R), green (G) and blue (B) data, so color information is usually generated using surrounding pixels. The method used to generate green is especially important because of the major contribution made by green to resolution. In the ClearVid CMOS Sensor, each pixel of another color is completely surrounded by green pixels. This arrangement is suitable for enhancing resolution.

Image signal output from the ClearVid CMOS Sensor is processed by the Enhanced Imaging Processor, using an original algorithm optimized for the sensor's unique pixel array. This image processing engine can produce either HD video or high resolution still images. The development of the image processing algorithm involved exhaustive testing to ensure that it would be able to exploit the full potential of the ClearVid CMOS Sensor. With these innovations to the pixel array and imaging processing engine, Sony has created a system that uses the same number of recording pixels as conventional sensors but provides significantly better sensitivity and equivalent resolution.

Sony's ClearVid CMOS Sensor has been used in numerous Handycam models, especially compact HD video cameras. The first to feature the new sensor was the HDR-HC3, a digital HD Handycam launched in 2006. By partnering the ClearVid CMOS Sensor with the Enhanced Imaging Processor, Sony has been able to create unique capabilities, including simultaneous recording of HD video and high-resolution still images, and high-speed shooting at 240fps. Furthermore, because high-quality images are created using less readout pixels than conventional sensors, Sony has also made an important contribution to the development of highly compact products and reducing power consumption.

Starting with HDR-SR11, which will be launched in the spring of 2008, Sony plans to combine 1/3.13 ClearVid CMOS Sensor with Exmor noise reduction technology used in Sony's α700 digital SLR camera and other products. In addition to its "column AD conversion" function, which carries out analog-digital conversion in each column on the CMOS sensor, Exmor also eliminates noise from the initial stage of signal conversion, thereby helping to minimize noise even at high sensitivity. By combining this technology with its new BIONZ image processing engine, Sony has created a system capable of recording high quality still and video images with high resolution, high sensitivity and low noise. Because this system produces approximately 3.6 times more image data (3,680x2,070) than a 1920x1080i system, this system can record bright, high resolution images.