| 摘要 |
Systems and devices for processing image or other data using product-law symmetry are described. In one implementation, an image or other collection of data has a number of samples each having an associated intensity, luminance, magnitude or other value. A sub-set of the samples is selected, and a product is computed of at least a first value associated with a first sample in the sub-set and a second value associated with a second sample in the sub-set that is different from the first sample. The resulting product can be used to provide an output, such as an enhanced image or an indication that a target is present within the sub-set of samples. In image processing applications, the product may be based upon data obtained from single or multiple images to exploit target symmetry and to distribute the effects of random noise, thereby improving target identification or otherwise enhancing the image.
|
