Super-resolution imaging for a multi-sensor internet of things environment
Abstract:
Image combination techniques are provided for a multi-sensor Internet of Things environment. An exemplary method comprises: dynamically determining an image resolution for at least a portion of an image to be collected by and/or transmitted by a plurality of image sensors within a distributed network based on one or more predefined image resolution rules with respect to an available bandwidth; and combining the image portions from the plurality of image sensors to generate at least one higher resolution image, such as a super-resolution image. The predefined image resolution rules specify, for example, that the portions of an image that have changed should be transmitted with a higher resolution relative to portions of the image that have not changed; and/or that multiple versions of a given image should be combined when the given image is degraded by noise. A plurality of low resolution images having a substantially similar alignment are optionally obtained and combined to obtain at least one high resolution image.
Information query
Patent Agency Ranking
0/0