Fixed pattern noise removal method

1. A fixed pattern noise (FPN) removal method for an image sensor, comprising:

calculating each compensation value corresponding to each pixel column according to a plurality of compensation pixel values of the each pixel column; and

sampling a plurality of active pixel values in an active pixel area and compensating the plurality of active pixel values according to the each compensation value of the each pixel column, to generate a plurality of compensated active pixel values;

wherein a plurality of active pixels sense light to generate the plurality of active pixel values;

wherein the step of calculating each compensation value corresponding to each pixel column according to the plurality of compensation pixel values of the each pixel column comprises:

setting a specific operational gain; and

sampling and calculating a difference between each primary color pixel column average of pixels of each primary color in the each pixel column in an optical black area and each primary color pixel average of pixels of the each primary color in the optical black area under the specific operational gain and an initial state, and saving the difference to a memory;

wherein all pixels in the optical black area do not sense light.

2. The FPN removal method of claim 1 , wherein the step of calculating the each compensation value corresponding to the each pixel column according to the plurality of compensation pixel values of the each pixel column comprises:

calculating each optical black area specific primary color pixel average of pixels of all specific primary colors in the optical black area under an analog gain; and

multiplying the difference between the each primary color pixel column average of the pixels of the each primary color in the each pixel column in an optical black area and the each primary color pixel average of the pixels of the each primary color in the optical black area by a ratio of the analog gain to the specific operational gain, and then summing up the each optical black area specific primary color pixel average as the each compensation value of the each pixel column.

3. The FPN removal method of claim 1 , wherein the specific operational gain is a maximum operational gain.

4. The FPN removal method of claim 1 , wherein the step of calculating the each compensation value corresponding to the each pixel column according to the plurality of compensation pixel values of the each pixel column comprises:

sampling and calculating each optical black area pixel column average of the each pixel column in an optical black area as the each compensation value of the each pixel column when the plurality of active pixels in the active pixel area sense light.

5. The FPN removal method of claim 1 , wherein the step of calculating the each compensation value corresponding to the each pixel column according to the plurality of compensation pixel values of the each pixel column comprises:

cutting off connections between a sampler and the plurality of active pixels in the active pixel area;

and calculating a difference between each primary color circuit column average of pixels of each primary color in the each pixel column in the active pixel area corresponding to the sampler and each primary color circuit average of pixels of the each primary color in the active pixel area corresponding to the sampler under the specific operational gain and an initial state, and saving the difference to a memory.

6. The FPN removal method of claim 1 , wherein the step of calculating the each compensation value corresponding to the each pixel column according to the plurality of compensation pixel values of the each pixel column comprises:

calculating an optical black area pixel average of all pixels in an optical area; and

multiplying the difference between the each primary color circuit average of the pixels of the each primary color in the active pixel area corresponding to the sampler and the each primary color circuit column average of the pixels of the each primary color in the each pixel column in the active pixel area corresponding to the sampler by a ratio of an analog gain to the specific operational gain, and then summing up the optical black area pixel average as the each compensation value of the each pixel column;

wherein all pixels in the optical black area do not sense light.

7. The FPN removal method of claim 5 , wherein the specific operational gain is a maximum operational gain.

8. The FPN removal method of claim 1 , wherein the each compensation value and the plurality of active pixel values are compensated in a digital form.

9. The FPN removal method of claim 1 , wherein the each compensation value and the plurality of active pixel values are compensated in an analog form.