Pixim, Incorporated

Benjamin P Olding

Justin Reyneri

William R Bidermann

Michael G Frank

Ricardo Jansson Motta

An image capture method in a digital image sensor implements a continuous sampling technique with a massively parallel thermometer-code analog-to-digital conversion (ADC) technique to generate pixel data having an intrinsic to an ultra-high dynamic range. In one embodiment, the method includes, after an initial exposure period, sampling pixel intensity values at the pixel elements of an image sensor at multiple sampling intervals within a snapshot of a scene; providing an analog reference voltage corresponding to a decrementing digital count value; and comparing the pixel intensity values to the analog reference voltage at each of the multiple sampling intervals. If the pixel intensity value of a first pixel element is less than the analog reference voltage at a first exposure time, an output signal having a first value is generated and the digital count value is recorded as pixel data associated with the first pixel element.

Digital image capture having an ultra-high dynamic range

An image capture method in a digital image sensor implements a continuous sampling technique with a massively parallel thermometer-code analog-to-digital conversion (ADC) technique to generate pixel data having an intrinsic to an ultra-high dynamic range. In one embodiment, the method includes, after an initial exposure period, sampling pixel intensity values at the pixel elements of an image sensor at multiple exposure times within a snapshot of a scene; providing an analog reference voltage corresponding to a decrementing digital count value; and comparing the pixel intensity values to the analog reference voltage at each of the multiple exposure times. If the pixel intensity value of a first pixel element is less than the analog reference voltage at a first exposure time, an output signal having a first value is generated and the digital count value is recorded as pixel data associated with the first pixel element.

An image sensor includes a two-dimensional array of pixel elements where the array of pixel elements outputting pixel data representing an image of a scene, and a two-dimensional array of selectively transmissive filters superimposed on the two-dimensional array of pixel elements, whereby each pixel element in the array of pixel elements is disposed to capture a first and a second color spectra of visible light. In one embodiment, the image sensor is a digital pixel sensor where the array of pixel elements is a sensor array of digital pixels, each of the digital pixels outputting digital signals as pixel data. In another embodiment, the pixel elements of the image sensor output analog signals as pixel data. In this manner, light intensity values for two different color spectra are optically summed at the pixel level, providing pixel values that are suitable for use in interlaced video display. The image sensor of the present invention has applications in video cameras employing the CMYG (cyan, magenta, yellow, and green) interlaced capture scheme for capturing full color video images.

Image sensor and method for implementing optical summing using selectively transmissive filters

A method for storing image information in a digital pixel sensor is disclosed for reducing the size of the memory needed to facilitate multiple sampling. An image sensor includes a sensor array of pixel elements generating pixel data in k bits and a data memory for storing pixel data of each pixel element. In one embodiment, the data memory in the digital pixel sensor stores an m-bit time index value and the lower k&#x2212;1 bits of the pixel data for pixel data exceeding the predetermined threshold level in all exposure times except the last. When pixel data in the last exposure time is captured, the data memory stores the time index value in less than m bits and stores the k-bit pixel data. The size of the data memory is minimized while preserving the image quality. The method of the present invention provides the benefits of reducing manufacturing cost and improving production yield.

Method and apparatus for storing image information for multiple sampling operations in a digital pixel sensor

A method for generating coefficients for a set of convolution kernels for use in interpolating pixel values in an image sensor is described. The coefficients are computed by applying a constraint matrix specifying one or more constraints. The method includes generating ideal sensor data representative of a test image in a first color plane, generating sensor data of the test image, generating f data matrices including pixel data from multiple neighborhoods of pixels in the pixel array, and determining the coefficients for f convolution kernels using the ideal sensor data, the f data matrices and by applying one or more constraints. The use of a constraint matrix greatly simplifies the computation of the coefficients and can be applied in image processing to generate a high quality full color image.

Method of defining coefficients for use in interpolating pixel values

A method for storing image information in a digital pixel sensor is disclosed for reducing the size of the memory needed to facilitate multiple sampling and correlated double sampling. In one embodiment, an image sensor includes a sensor array of pixel elements generating digital pixel data and a data memory for storing pixel data of each pixel element. The data memory allocates for each pixel element an m-bit pixel data field for storing pixel data and an n-bit CDS data field for storing reset value associated with each pixel element. The data memory stores m+n bits of pixel data in the pixel data field and the CDS data field for pixel data exceeding the predetermined threshold level. The data memory stores m bits of pixel data in the pixel data field for pixel data not exceeding the predetermined threshold level.

Growing community of inventors

Redwood City, CA, United States of America

Benjamin P Olding