AS0142AT datasheet(15/45 Pages) ONSEMI

15 / 45 page

AS0142AT

www.onsemi.com

ADACD (ADAPTIVE COLOR DIFFERENCE)

The next step in the image stream process is noise

reduction. The AS0142AT uses a noise reduction filter

called AdaCD which focuses on removing color noise while

preserving edge details. Automotive applications require

good performance in extremely low light

, even at high

temperature conditions. In these stringent conditions the

image sensor is prone to higher noise levels

. Efficient noise

reduction techniques are required to circumvent this sensor

limitation and deliver a high quality image.

BLACK LEVEL SUBTRACTION AND DIGITAL GAIN

After noise reduction

, the pixel data goes through black

level subtraction and multiplication by a programmable

digital gain

. This digital gain can be programmed

independently per color channel registers. Black level

subtraction (to compensate for sensor data pedestal) is a

single value applied to all color channels. If the black level

subtraction produces a negative result for a particular pixel,

the value of this pixel is set to 0.

POSITIONAL GAIN ADJUSTMENTS (PGA)

Lens has tendency to produce images whose brightness is

significantly attenuated near the edges

. On top of other

factors causing fixed pattern signal gradients in images

captured by image sensors, image shading occurs

. The

AS0142AT has an embedded shading correction module

that can be programmed to counter the shading effects on

each individual R

, Gb, Gr, and B color signal.

THE CORRECTION FUNCTION

The correction functions can then be applied to each pixel

value to equalize the response across the image as per

following relation:

Pcorrected(row, col) + Psensor(row, col) f(row, col)

(eq. 1)

where P is the pixel value and f is the color dependent

correction functions for each color channel

ADAPTIVE LOCAL TONE MAPPING (ALTM)

Real world scenes often have very high dynamic range

(HDR) that far exceeds the electrical dynamic range of the

imager. Dynamic range is defined as the luminance ratio

between the brightest and the darkest object in a scene. In

recent years many technologies have been developed to

capture the full dynamic range of real world scenes. For

example

, the multiple exposure method is widely adopted

for capturing high dynamic range images

, which combines

a series of low dynamic range images of the same scene

taken under different exposure times into a single HDR

image.

Even though the new digital imaging technology enables

the capture of the full dynamic range, low dynamic range

display devices can be the limiting factor. Typical LCD

monitor nowaday has contrast ratio around 1,000:1 which is

not enough for an HDR image (the contrast ratio for an HDR

image is around 250,000:1). Therefore, in order to

reproduce HDR images on a low dynamic range display

device

, the captured high dynamic range must be

compressed to the available range of the display device. This

is commonly called tone mapping.

Tone mapping methods can be classified into global tone

mapping and local tone mapping. Global tone mapping

methods apply the same mapping function to all pixels.

While

global

tone

mapping

methods

provide

computationally simple and easy to use solutions

, they often

cause loss of contrast and detail

. A local tone mapping is thus

necessary in addition to global tone mapping

, for the

reproduction of visually higher appealing images that could

reveal scene details which are important for automotive

safety and surveillance applications. Local tone mapping

methods use a spatially variable mapping function

determined by the neighborhood of a pixel, which allows it

to increase the local contrast and the visibility of some

details of the image. Local methods usually yield more

pleasing results because they exploit the fact that human

vision is more sensitive to local contrast.

ON Semiconductor’s ALTM solution significantly

improves the performance over global tone mapping.

ALTM is directly applied to the Bayer domain to compress

the dynamic range from 20

-bit to 12−bit. This allows the

regular color pipeline to be used for HDR image rendering

COLOR INTERPOLATION

In the raw data stream fed by the embedded sensor to the

IFP, each pixel is represented by a 20− or 12−bit integer

number, which can be considered proportional to the pixel’s

response to a one−color light stimulus, red, green, or blue,

depending on the pixel’s position under the color filter array.

Initial data processing steps

, up to and including ALTM,

preserve the one−color−per−pixel nature of the data stream,

but after ALTM it must be converted to a

three−colors−per−pixel stream appropriate for standard

color processing. The conversion is done by an edge

sensitive color interpolation module. The module pads the

incomplete color information available for each pixel with

information extracted from an appropriate set of

neighboring pixels. The algorithm used to select this set and

extract the information seeks the best compromise between

preserving edges and filtering out high frequency noise in

flat field areas

. The edge threshold can be set through

COLOR CORRECTION AND APERTURE CORRECTION

To achieve good color fidelity of the IFP output

interpolated RGB values of all pixels are subjected to color

correction

. The IFP multiplies each vector of three pixel

colors by a 3 x 3 color correction matrix

. The three

components of the resulting color vector are all sums of three

10−bit numbers. The color correction matrix (CCM) can be

either programmed by the user or automatically selected by

the auto white balance (AWB) algorithm implemented in the

IFP

. Color correction should ideally produce output colors

that are corrected for the spectral sensitivity and color

AS0142AT Datasheet(PDF) 15 Page - ON Semiconductor

AS0142AT Datasheet(HTML) 15 Page - ON Semiconductor

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Datasheet Download

Link URL