Vision recognition apparatus and method
AssigneeSamsung Electronics Co., Ltd.
InventorsHwa-Yong Kang, Young-Kwon Yoon, Young-Sam Yu, Eun-Soo Chang
Life of this patent
Estimating the term…
Abstract
Some of pixels in a general image sensor for image capturing are designated as vision pixels used for vision recognition. Optical information obtained only from the vision pixels is used to perform vision recognition in the vision recognition mode. Capturing image data is generated based on optical information obtained from all the pixels in the image data in the image-shooting mode. A digital signal converter that converts optical information obtained only from vision pixels into a digital signal is separately provided in addition to a digital signal converter that converts optical information of all pixels in the image sensor into a digital signal.
Claims15
A vision recognition apparatus in a digital device, comprising:
- an image sensor including a plurality of pixels, a portion of the plurality of pixels being designated as vision pixels to be used for vision recognition;
- a first digital signal converter configured to digitally convert a signal output from the designated vision pixels;
- a second digital signal converter configured to digitally convert a signal output from the plurality of pixels;
- and
- a processor configured to:
- generate recognition image data to be used for vision recognition by using optical information obtained from the designated vision pixels by processing signals output from the first digital signal converter in a vision recognition mode,
- perform vision recognition by identifying and detecting a specific object included in the generated recognition image data, and
- generate capturing image data by processing signals output from the second digital signal converter in a shooting mode,
- wherein a number of the vision pixels to be used for vision recognition is determined depending on a resolution of the recognition image data,
- wherein a resolution of the vision pixels to be used for vision recognition is less than a resolution of other pixels in the plurality of pixels,
- wherein the vision pixels have higher sensitivities among the plurality of pixels having different sensitivities, the vision pixels including white pixels, and
- wherein the capturing image data is generated based on results obtained by performing pixel interpolation on optical information of the vision pixels.
A vision recognition method in a digital device with an image sensor including a plurality of pixels, the method comprising:
- determining whether an operation mode is a vision recognition mode;
- generating, if the operation mode is a vision recognition mode, recognition image data to be used for vision recognition by using optical information obtained from vision pixels by processing signals output from a first digital signal converter that digitally converts a signal output from a portion of the plurality of pixels being designated as the vision pixels to be used for vision recognition in the image sensor;
- performing vision recognition by identifying and detecting a specific object included in the generated recognition image data;
- and
- generating, if the operation mode is a shooting mode, capturing image data by processing signals output from pixels selected from among the plurality of pixels depending on a specific resolution, by a second digital signal converter that digitally converts a signal output from each of the plurality of pixels,
- wherein a number of the vision pixels to be used for vision recognition is determined depending on a resolution of the recognition image data,
- wherein a resolution of the vision pixels to be used for vision recognition is less than a resolution of other pixels in the plurality of pixels,
- wherein the vision pixels have higher sensitivities among the plurality of pixels having different sensitivities, the vision pixels including white pixels, and
- wherein the capturing image data is generated based on results obtained by performing pixel interpolation on optical information of the vision pixels.
A low-power vision recognition method in a mobile terminal with an image sensor including a plurality of pixels, the method comprising:
- receiving an image from the image sensor;
- extracting a recognition image input to a portion of the plurality of pixels being designated as vision pixels to be used for vision recognition in the image sensor;
- and
- performing vision recognition by identifying and detecting a specific object included in the recognition image input to the vision pixels,
- wherein the extracting comprises generating, if an operation mode of the mobile terminal is a vision recognition mode, recognition image data to be used for vision recognition by using optical information obtained from the designated vision pixels by digitally converting a signal output from the portion of the plurality of pixels, and
- wherein a number of the vision pixels to be used for vision recognition is determined depending on a resolution of the recognition image data,
- wherein a resolution of the vision pixels to be used for vision recognition is less than a resolution of other pixels in the plurality of pixels,
- wherein the vision pixels have higher sensitivities among the plurality of pixels having different sensitivities, the vision pixels including white pixels, and
- wherein the capturing image data is generated based on results obtained by performing pixel interpolation on optical information of the vision pixels.
Dependencies are inferred from the claim text.