Methods of sensor mode switching in event based sensor and imaging camera for low power application
AssigneeSony Semiconductor Solutions Corporation
InventorsSa Xiao, Kevin Chan, Ping Wah Wong
Life of this patent
Estimating the term…
Abstract
An image sensor comprises a plurality of pixels. Pixels are capable of detecting a change in an amount of light intensity and pixels are capable of detecting an amount of light intensity. In a first mode the sensor outputs data from the first one or more of the pixels. In a second mode the sensor outputs data from the second one or more of the pixels. At least one of the first mode and the second mode is selected by a processor based on at least one of a result of processing data output in the first mode and a result of processing data output in the second mode.
Claims20
A system, comprising:
- a first sensor configured to detect a change in light intensity;
- a second sensor configured to detect an amount of light intensity;
- an output circuit;
- and
- a processor configured to:
- process data output by the first sensor using a neural network,
- detect, based on processing the data using the neural network, an object in the data output by the first sensor based on an event density in the data output by the first sensor exceeding or falling below a threshold,
- determine a frame rate based on a result of the neural network processing the data output by the first sensor, wherein the frame rate is based at least in part on an identification of the object, and
- in response to detecting the object, cause the second sensor to capture data at the determined frame rate and switch an output of output circuit from an output of the first sensor to an output of the second sensor.
An imaging system, comprising:
- a first sensor configured to detect a change in light intensity;
- and
- a second sensor configured to detect an amount of light intensity, wherein:
- a neural network processes data output by the first sensor,
- a processor detects, based on an output of the neural network, an object in the data output by the first sensor based on an event density in the data output by the first sensor exceeding or falling below a threshold,
- the processor determines a frame rate based on a result of the neural network processing the data output by the first sensor, wherein the frame rate is based at least in part on an identification of the object,
- the processor causes the second sensor to capture data at the determined frame rate and switches an output of an output circuit from an output of the first sensor to an output of the second sensor in response to detecting the object.
An image sensor comprising a plurality of pixels, wherein
- a first one or more of the pixels are capable of detecting a change in light intensity,
- a second one or more of the pixels are capable of detecting an amount of light intensity,
- a neural network processes data output by the first one or more of the pixels,
- a processor detects, based on an output of the neural network, an object in the data output by the first one or more of the pixels based on an event density in the data output by the first one or more pixels exceeding or falling below a threshold,
- the processor determines a frame rate based on a result of the neural network processing the data output by the first one or more pixels, wherein the frame rate is based at least in part on an identification of the object,
- in response to detecting the object, the processor causes the second one or more of the pixels to capture data at the determined frame rate the processor switches an output of an output circuit from an output of the first sensor to an output of the second sensor.
Dependencies are inferred from the claim text.