System and method for handling bad pixels in image sensors

Abstract

A method and apparatus for handling bad pixels in an image sensor array includes processing data values associated with the pixels of the image sensor. Processing the data values is performed by at least a first pass process and a subsequent pass process. In the first pass, the data values associated with the pixels are analyzed to determine whether any of the pixels are bad pixels. Information identifying the bad pixels is stored in a memory storage area of limited size. The stored information may also include an indicator, indicating a confidence level in categorizing the bad pixel. During the first pass, an overflow mark is stored in the memory storage area when insufficient memory storage is available for storing the information about a particular bad pixel. The overflow mark identifies the particular pixel in the image sensor array. During subsequent passes, processing may be redirected to the first pass to resume first pass processing of the data value associated with the particular pixel identified by the overflow mark. Based on information stored in the memory storage area, subsequent passes determine which data values are re-evaluated and whether the categorization of any particular bad pixel may be modified based on the re-evaluation. Corrected data values for each bad pixel may be determined by interpolating data values associated with neighboring pixels to each bad pixel. The first pass may be performed in response to a reset signal and the subsequent pass may be performed at a pre-determined time interval or event.