Identifying contiguous regions of constant pixel intensity in images
A technique identifies regions of an image characterized by constant pixel intensity in a resource-efficient, latency-efficient, and scalable manner. The technique involves: obtaining a candidate image; determining whether the candidate image contains a contiguous region of pixels having intensity values within a specified range of intensity values; assessing whether the contiguous region satisfies a prescribed test; and selecting or excluding the candidate image for further processing based on a result of the assessing. The operation of determining involve two phases. First, the technique determines a distribution of intensity values within the candidate image. Second, the technique leverages the distribution to search the candidate image for neighboring pixels having intensity values within the specified range of intensity values, beginning from a selected starting pixel in a qualifying subset of pixels. In some examples, the technique is applied to the task of combining supplemental content with the candidate image.
1 . A computer-implemented method for processing images, comprising:
obtaining a candidate image;
generating a distribution of pixel intensity values in the candidate image, each of the pixel intensity values expressing an intensity of a corresponding pixel in the candidate image;
in a portion analysis process,
determining a quantity of pixels in an identified portion of the distribution, the identified portion demarcating a range of pixel intensity values in the distribution;
determining whether the quantity of pixels in the identified portion satisfies a prescribed threshold value;
upon determining that the quantity of pixels satisfies the prescribed threshold value, determining whether the candidate image contains a contiguous region of pixels having pixel intensity values within the range of pixel intensity values;
upon determining that the candidate image contains the contiguous region, assessing whether the contiguous region satisfies a prescribed test, wherein the test includes determining whether the contiguous region intersects at least one edge of the candidate image by a prescribed amount; and
selecting or excluding the candidate image for further processing based on a result of the assessing.
2 . The computer-implemented method of claim 1 , wherein the distribution is expressed as a histogram that specifies counts of pixels for respective subranges of pixel intensity values, and wherein the identified portion encompasses one or more subranges.
3 . The computer-implemented method of claim 1 , wherein the method involves performing the portion analysis process for at least one other portion and associated range of the distribution.
4 . The computer-implemented method of claim 1 , wherein the determining whether the quantity of pixels in the identified portion satisfies a prescribed threshold value involves determining whether the quantity of pixels relative to a total quantity of pixels in the candidate image satisfies the prescribed threshold value.
5 . The computer-implemented method of claim 1 , wherein the determining whether the candidate image contains a contiguous region of pixel intensity values involves performing a search for neighboring pixels in the candidate image with pixel intensity values within the range of pixel intensity values, starting from a selected starting pixel.
6 . The computer-implemented method of claim 5 , wherein the search is a breadth-first search.
7 . The computer-implemented method of claim 1 , further including sorting pixels in the identified portion with respect to location of the pixels in the identified portion across the candidate image, with respect to at least one axis of the candidate image, to produce a sorted set of pixels.
8 . The computer-implemented method of claim 7 , wherein the determining whether the candidate image contains a contiguous region of pixel intensity values is performed for a specified subset of pixels in the sorted set of pixels.
9 . The computer-implemented method of claim 8 , wherein a search for pixels having pixel intensity values within the range of pixel intensity values is commenced from a starting pixel in the subset of pixels upon determining that the pixels in the subset of pixels are spatially-connected pixels and share a same position on one axis of the candidate image.
10 . The computer-implemented method of claim 1 , wherein the test involves also includes determining whether a quantity of pixels in the contiguous region satisfies another prescribed threshold value, with respect to a total quantity of pixels in the candidate image.
11 . The computer-implemented method of claim 1 , wherein the further processing involves combining the candidate image with supplemental content.
12 . The computer-implemented method of claim 11 , further including serving the candidate image and the supplemental content as an advertisement.
13 . The computer-implemented method of claim 1 , further including generating a first value that indicates whether the candidate image contains the contiguous region, and a second set of values that identify the pixels in the contiguous region.
14 . A computing system, comprising:
a processing system including a processor; and
a storage device for storing machine-readable instructions,
the processing system executing the machine-readable instructions in the storage device to perform operations comprising:
obtaining a candidate image;
determining whether the candidate image contains a contiguous region of pixel intensity values within a specified range of intensity values,
the determining involving determining a distribution of pixel intensity values within the candidate image, and, guided by the distribution of pixel intensity values, searching the candidate image for neighboring pixels having pixel intensity values within the specified range of pixel intensity values;
upon determining that the candidate image contains the contiguous region, assessing whether the contiguous region satisfies a prescribed test, wherein the test also involves determining whether the contiguous region intersects with at least one edge of the candidate image by a prescribed amount; and
selecting or excluding the candidate image for further processing based on a result of the assessing.
15 . The computing system of claim 14 , wherein the test involves also includes determining whether a quantity of pixels in the contiguous region satisfies a prescribed threshold value, with respect to a total quantity of pixels in the candidate image.
16 . The computing system of claim 14 , wherein the operations further include generating a first value that indicates whether the candidate image contains the contiguous region, and a second set of values that identify the pixels in the contiguous region.
17 . A computer-readable storage medium for storing computer-readable instructions, wherein a processing system executes the computer-readable instructions to perform operations comprising:
obtaining a candidate image;
generating a distribution of pixel intensity values in the candidate image, each of the pixel intensity values expressing a pixel intensity of a corresponding pixel in the candidate image;
determining a quantity of pixels in an identified portion of the distribution, the identified portion demarcating a range of pixel intensity values in the distribution;
sorting pixels in the identified portion with respect to location of the pixels in the identified portion across the candidate image, with respect to at least one axis of the candidate image, to produce a sorted set of pixels;
determining whether the candidate image contains a contiguous region of pixel intensity values having intensity values within the range of intensity values, wherein the determining whether the candidate image contains a contiguous region of pixel intensity values is performed for a specified subset of pixels in the sorted set of pixels, and wherein a search for pixels having pixel intensity values within the range of pixel intensity values is commenced from a starting pixel in the subset of pixels upon determining that the pixels in the subset of pixels are spatially-connected pixels and share a same position on one axis of the candidate image; and
upon determining that the candidate image contains the contiguous region, assessing whether the contiguous region satisfies a prescribed test.
18 . The computer-implemented method of claim 1 , further comprising repeating the portion analysis process for another identified portion of the distribution.