An image sensor may include an array of pixels arranged in rows and columns. Each pixel can include a subpixel circuit and one or more subtraction circuits. Each subpixel circuit can be selectively coupled to neighboring subpixel circuits in the same row via horizontal odd and even switches and can be selectively coupled to neighboring subpixel circuits in the same column via vertical odd and even switches. The horizontal and vertical switches can be turned on in separate phases to store difference values from pairs of neighboring subpixels into the one or more subtraction circuits. The difference values can be read out using comparators and one or more shift registers to output an edge image that includes only edge information.

An image sensor may include an array of pixels arranged in rows and columns. Each pixel can include a subpixel circuit and one or more subtraction circuits. Each subpixel circuit can be selectively coupled to neighboring subpixel circuits in the same row via horizontal odd and even switches and can be selectively coupled to neighboring subpixel circuits in the same column via vertical odd and even switches. The horizontal and vertical switches can be turned on in separate phases to store difference values from pairs of neighboring subpixels into the one or more subtraction circuits. The difference values can be read out using comparators and one or more shift registers to output an edge image that includes only edge information.

A CMOS imaging sensor with embedded feature extraction capability operatable in different modes by a method that includes the steps of: (a) operating the CMOS imaging sensor in a motion-detecting mode at a first power level using circuitry on the imaging sensor that generates motion data based on received images detected by pixels in the pixel array; (b) switching the imaging sensor from the motion-detecting mode to a feature extraction mode in response to detecting motion; and (c) operating the imaging sensor in the feature extraction mode at a second power level that is higher than the first power level.

Embodiments of the technology disclosed herein provide filtering pixels for performing filtering of signals within the pixel domain. Filtering pixels described herein include a hybrid analog-digital filter within the pixel circuitry, reducing the need for computationally intensive and inefficient post-processing of images to filter out aspects of captured signals, and account for ineffectiveness of optical filters.

A dual mode image sensor is provided. The image sensor includes an on-chip sensing array, on-chip analog-to-digital converters, and an on-chip processor. The sensor array has rows and columns of discrete sensor elements. The dual mode image sensor has a scene sensing mode and an image capture mode, which use the same set of imaging optics. The processor includes a dual context register; one being for the scene sensing mode and the other for image capture mode. The scene sensing mode is configured to output results of object sensing, motion detection, focus evaluation and illumination measurement to the analog-to-digital converters. The image capture mode is configured to output captured images to the analog-to-digital converters, which are configured to send the digital data to the processor. The processor is configured to switch from scene sensing mode to image capture mode based upon the scene sensing mode output results.

To provide an imaging device capable of reducing the amount of data with a simple method. An imaging device includes: a plurality of sensor elements which is arranged in a matrix shape and each of which generates a photoelectric conversion voltage in accordance with an input light level; and a read circuit which is coupled to bit lines provided while being associated with respective columns of the sensor elements, and amplifies and reads the photoelectric conversion voltages generated in the sensor elements by being exposed at predetermined timing. The read circuit outputs differential data of the read photoelectric conversion voltages generated in the respective sensor elements that are adjacent to each other in the same row or column.

Provided is an imaging device operated at high speed and low power consumption. The imaging device includes a pixel and a first circuit. The pixel includes a first photoelectric conversion element and a second photoelectric conversion element. The first circuit is configured to compare a first signal which is output from the pixel on the basis of imaging data obtained by the first photosensitive conversion element to a second signal which is output from the pixel on the basis of imaging data obtained by the second photosensitive conversion element for determining whether there is a difference between the first signal and the second signal. Thus, edge detection can be performed without a periphery device for edge detection outside the imaging device.

Certain aspects relate to systems and techniques for using imaging pixels (that is, non-phase detection pixels) in addition to phase detection pixels for calculating autofocus information. Imaging pixel values can be used to interpolate a value at a phase detection pixel location. The interpolated value and a value received from the phase difference detection pixel can be used to obtain a virtual phase detection pixel value. The interpolated value, value received from the phase difference detection pixel, and the virtual phase detection pixel value can be used to obtain a phase difference detection signal indicating a shift direction (defocus direction) and a shift amount (defocus amount) of image focus.

A color filter array includes a plurality of first pixels transmitting light having visible wavelengths, a plurality of second pixels transmitting light having a first range of wavelengths in the visible spectrum, a plurality of third pixels transmitting light having a second range of wavelengths in the visible spectrum, and a plurality of crosstalk detection pixels each including a first half pixel and a second half pixel, wherein the first half pixel transmits light having first wavelengths in a part of the first range, and wherein the second half pixel partially transmits light having second wavelengths in a part of the second range.

Methods, systems, computer-readable storage media, and apparatuses for efficiently computing a computer vision operation are presented. In certain aspects, techniques are disclosed for receiving values from one or more pixels from a pixel array and representing those values for facilitating computer vision operations, such as Haar-like feature computations. In one implementation, the pixel values are represented as a hierarchy of computations; each level above the lower level of the hierarchy of computations comprises one or more values representing a computation of the values from a lower level of the hierarchy. The computation can include a simple sum of the values for the lower level of the hierarchy, a weighted sum of the values for the lower level of the hierarchy, an average of the values for the lower level of the hierarchy, a local binary pattern based on the values for the lower level of the hierarchy, a histogram of oriented gradients based on the values for the lower level of the hierarchy, or other computer vision feature computation based on the values for the lower level of the hierarchy. In such an implementation, the computer vision operations, such as Haar-like feature computations, may be performed using the hierarchy of averages.