A second substrate including a pixel circuit that outputs a pixel signal on a basis of electric charges outputted from the sensor pixel and a third substrate including a processing circuit that performs signal processing on the pixel signal are provided. The first substrate, the second substrate, and the third substrate are stacked in this order. A semiconductor layer including the pixel circuit is divided by an insulating layer. The insulating layer divides the semiconductor layer to allow a center position of a continuous region of the semiconductor layer or a center position of a region that divides the semiconductor layer to correspond to a position of an optical center of the sensor pixel, in at least one direction on a plane of the sensor pixel perpendicular to an optical axis direction.

Provided are an information processing device, an information processing method, and an information processing program capable of reducing a processing load of convolution processing in a convolutional neural network (CNN). An information processing device (1) according to the present disclosure includes a setting unit (51) and a control unit (52). The setting unit (51) sets exposure time of each of imaging pixels in an imaging unit (2), which includes a plurality of imaging pixels arrayed two-dimensionally, to exposure time corresponding to a convolution coefficient of a first layer of a CNN. The control unit (52) causes transfer of signal charges from imaging pixels, which have been exposed, to a floating diffusion (FD), thereby performing convolution processing.

Embodiments of the present disclosure are directed to an image sensor. The image sensor includes a color filter array, a pixel array, and a plurality of analog-to-digital converters (ADCs). The ADCs convert the analog pixel signal obtained by pixels corresponding to first color filters into a digital pixel signal based on a first bit precision and to convert the analog pixel signal obtained by pixels corresponding to second color filters and third color filters into a digital pixel signal based on a second bit precision. The second bit precision is lower than the first bit precision.

An image sensor and a method of operating the same are provided. The image sensor includes a semiconductor substrate of a first conductivity type; a photoelectric conversion region provided in the semiconductor substrate and doped to have a second conductivity type; a first floating diffusion region provided to receive photocharges accumulated in the photoelectric conversion region; a transfer gate electrode disposed between and connected to the first floating diffusion region and the photoelectric conversion region; a dual conversion gain transistor disposed between and connected to the first floating diffusion region and a second floating diffusion region; and a reset transistor disposed between and connected to the second floating diffusion region and a pixel power voltage region, wherein a channel region of the reset transistor has a potential gradient increasing in a direction from the second floating diffusion region toward the pixel power voltage region.

An imaging system includes an illumination unit and a sensor unit disposed on a printed circuit board. The illumination unit includes a diode laser source inside an illumination housing. The sensor unit includes an image sensor having a pixel array and a lens barrel mounted on the image sensor with an adhesive, and an optical fiber coupled between the illumination housing and image sensor. The optical fiber is configured to collect a portion of light from the interior of the illumination housing that is emitted by the diode laser source and direct the portion of light to a corner of the pixel array of the image sensor that is located outside the lens barrel.

To improve the image quality of image data in a solid-state imaging device that reads a signal according to a potential difference between respective floating diffusion regions of a pair of pixels.

A pixel unit is provided with a plurality of rows each including a plurality of pixels. A readout row selection unit selects any of the plurality of rows as a readout row every time a predetermined period elapses, and causes each of the plurality of pixels in the readout row to generate a signal potential according to a received light amount. A reference row selection unit selects a row different from a previous row from among the plurality of rows as a current reference row every time the predetermined period elapses, and causes each of the plurality of pixels in the reference row to generate a predetermined reference potential. A readout circuit unit reads a voltage signal according to a difference between the signal potential and the reference potential.

A solid state imaging device includes a pixel array unit in which color filters of a plurality of colors are arrayed with four pixels of vertical 2 pixels×horizontal 2 pixels as a same color unit that receives light of the same color, shared pixel transistors that are commonly used by a plurality of pixels are intensively arranged in one predetermined pixel in a unit of sharing, and a color of the color filter of a pixel where the shared pixel transistors are intensively arranged is a predetermined color among the plurality of colors. The present technology can be applied, for example, to a solid state imaging device such as a back-surface irradiation type CMOS image sensor.

An image sensor includes a pixel portion in which a plurality of unit pixels each having one micro lens and a plurality of photoelectric conversion portions are arrayed in a matrix, a signal readout portion that reads out signals accumulated in the photoelectric conversion portions and converts the read signals to digital signals, a signal processor that processes signals read out by the signal readout portion and has an image capture signal processor that performs signal processing for generating a captured image on signals read out by the signal readout portion and a focus detection signal processor that performs signal processing for focus detection on signals read out by the signal readout portion, and an output portion that outputs signals processed by the signal processor.

A photoelectric conversion device includes: pixels forming columns and each configured to output a pixel signal; and comparator units provided to respective columns and each configured to receive the pixel signal from the pixels on a corresponding column and the reference signal. Each comparator unit includes a comparator having a first input node that receives the pixel signal and a second input node that receives the reference signal, a first capacitor that connects a reference signal line and the second input node, a second capacitor whose one electrode is connected to the second input node, and a select unit that connects the other electrode of the second capacitor to either the reference signal line or a reference voltage line. The other electrode of the second capacitor is connected to the reference signal line during first mode AD conversion and connected to the reference voltage line during second mode AD conversion.

A semiconductor apparatus includes a stack of a first chip having a plurality of pixel circuits arranged in a matrix form and a second chip having a plurality of electric circuit arranged in a matrix form. A wiring path between a semiconductor element configuring the pixel circuit and a semiconductor element configuring the electric circuit or a positional relationship between a semiconductor element configuring the pixel circuit and a semiconductor element configuring the electric circuit is differentiated among the electric circuits.