Learning device and method, recognition device and method, and program
A learning device includes: a generating unit configured to generate an image having different resolution from an input image; an extracting unit configured to extract a feature point serving as a processing object from an image generated by the generating unit; a calculating unit configured to calculate the feature amount of the feature point by subjecting the feature point to filter processing employing a predetermined filter; and an identifier generating unit configured to generate an identifier for detecting a predetermined target object from the image by statistical learning employing the feature amount; with the filter including a plurality of regions, and the calculating unit taking the difference value of difference within the regions as the feature amount.
1. A learning device comprising:
generating means configured to generate an image having different resolution from an input image;
extracting means configured to extract a feature point serving as a processing object from the image generated by the generating means;
calculating means configured to calculate a feature amount of the feature point by subjecting the feature point to filter processing employing a predetermined filter; and
identifier generating means configured to generate an identifier for detecting a predetermined target object from the generated image by statistical learning employing the feature amount;
wherein the predetermined filter includes a plurality of regions, and the calculating means takes a difference value of difference between a first region of the plurality of regions and a second region of the plurality of regions as the feature amount, wherein a size of the first region is different than a size of the second region, and
wherein the calculating means execute convolution operation in a differential function of a predetermined order of a Gaussian function in an arbitrary angle to calculate summation of absolute values of operation results.
2. The learning device according to claim 1 , wherein the plurality of regions of the predetermined filter have a rectangular shape.
3. A learning method comprising the steps of:
generating an image having different resolution from an input image;
extracting a feature point from the generated image;
calculating a feature amount of the feature point by subjecting the feature point to filter processing employing a predetermined filter; and
generating an identifier for detecting a predetermined target object from the generated image by statistical learning employing the feature amount;
wherein the predetermined filter includes a plurality of regions, and a difference value of the difference between a first region of the plurality of regions and a second region of the plurality of regions thereof is taken as the feature amount, wherein a size of the first region is different than a size of the second region, and
wherein convolution operation in a differential function of a predetermined order of a Gaussian function in an arbitrary angle is executed to calculate summation of absolute values of operation results.
4. A non-transitory computer-readable storage medium having stored thereon, a set of computer-executable instructions, the computer-executable instructions when executed by a computer causes the computer to perform steps comprising:
generating an image having different resolution from an input image;
extracting a feature point from the generated image;
calculating a feature amount of the feature point by subjecting the feature point to filter processing employing a predetermined filter; and
generating an identifier for detecting a predetermined target object from the generated image by statistical learning employing the feature amount;
wherein the predetermined filter includes a plurality of regions, and a difference value of the difference between a first region of the plurality of regions and a second region of the plurality of regions is taken as the feature amount, wherein a size of the first region is different than a size of the second region, and
wherein convolution operation in a differential function of a predetermined order of a Gaussian function in an arbitrary angle is executed to calculate summation of absolute values of operation results.
5. A learning device comprising:
a generating unit configured to generate an image having different resolution from an input image;
an extracting unit configured to extract a feature point serving as a processing object from the image generated by the generating unit;
a calculating unit configured to calculate a feature amount of the feature point by subjecting the feature point to filter processing employing a predetermined filter; and
an identifier generating unit configured to generate an identifier for detecting a predetermined target object from the generated image by statistical learning employing the feature amount;
wherein the predetermined filter includes a plurality of regions, and the calculating unit takes a difference value of difference between a first region of the plurality of regions and a second region of the plurality of regions as the feature amount, wherein a size of the first region is different than a size of the second region, and
wherein the calculating unit executes convolution operation in a differential function of a predetermined order of a Gaussian function in an arbitrary angle to calculate summation of absolute values of operation results.