Image driver that samples high-resolution image data
Techniques are disclosed for generating a low-dimensional representation of an image. An image driver receives an image captured by a camera. The image includes features based on pixel values in the image, and each feature describes the image in one or more image regions. The image driver generates, for each of the plurality of features, a feature vector that includes values for that feature corresponding to at least one of the image regions. Each value indicates a degree that the feature is present in the image region. The image driver generates a sample vector from each of the feature vectors. The sample vector includes each of the values included in the generated feature vectors.
1. A computer-implemented method, comprising:
generating, for each of a plurality of features of a first image in a sequence of images, a feature vector associated with a region of the first image and indicative of a degree to which that feature is present in the region of the first image;
generating a first sample vector based on the feature vectors, the first sample vector including values from each of the feature vectors; and
evaluating the first sample vector relative to at least a second sample vector associated with at least one further image in the sequence of images.
2. The computer-implemented method of claim 1 , wherein the sequence of images includes images captured by at least one of an infrared camera or a visible camera.
3. The computer-implemented method of claim 1 , further comprising sending the first sample vector to a sample memory for retrieval by a machine learning engine that evaluates the first sample vector.
4. The computer-implemented method of claim 1 , wherein the values from each of the feature vectors include at least one of: a color feature, an entropy feature, a homogeneity feature, or a gradient feature.
5. The computer-implemented method of claim 1 , further comprising:
receiving, at a first time, a profile specifying a spatial granularity for at least one feature from the plurality of features of the first image; and
receiving, at a second time after the first time, the first image.
6. The computer-implemented method of claim 1 , further comprising:
receiving a profile specifying a spatial granularity for at least one feature from the plurality of features of the first image; and
adjusting a non-zero number of regions of the first image based on the spatial granularity.
7. A non-transitory computer-readable medium storing instructions to cause a processor to:
generate, for each of a plurality of features of a first image in a sequence of images, a feature vector associated with a region of the first image and indicative of a degree to which that feature is present in the region of the first image;
generate a first sample vector based on the feature vectors, the first sample vector including values from each of the feature vectors; and
evaluate the first sample vector relative to at least a second sample vector associated with at least one further image in the sequence of images.
8. The computer-readable medium of claim 7 , wherein the sequence of images includes images captured by at least one of an infrared camera or a visible camera.
9. The computer-readable medium of claim 7 , further storing instructions to cause the processor to send the first sample vector to a sample memory for retrieval by a machine learning engine that evaluates the first sample vector.
10. The computer-readable medium of claim 7 , wherein the values from each of the feature vectors include at least one of: a color feature, an entropy feature, a homogeneity feature, or a gradient feature.
11. The computer-readable medium of claim 7 , further storing instructions to cause the processor to:
receive, at a first time, a profile specifying a spatial granularity for at least one feature from the plurality of features of the first image; and
receive, at a second time after the first time, the first image.
12. The computer-readable medium of claim 7 , further storing instructions to cause the processor to:
receive a profile specifying a spatial granularity for at least one feature from the plurality of features of the first image; and
adjust a number of regions of the first image based on the spatial granularity.
13. A computer-implemented method, comprising:
receiving a sequence of images, each image from the sequence of images including a plurality of features and each feature from the plurality of features being descriptive of at least one region of that image;
generating a sample vector based on a set of feature vectors associated with the plurality of features of a first image from the sequence of images, the sample vector including values from each feature vector from the set of feature vectors; and
evaluating the sample vector relative to at least one further sample vector associated with at least one further image in the sequence of images.
14. The method of claim 13 , wherein the sequence of images includes images captured by at least one of an infrared camera or a visible camera.
15. The method of claim 13 , further comprising sending the first sample vector to a sample memory for retrieval by a machine learning engine that evaluates the sample vector.
16. The method of claim 15 , wherein the machine learning engine evaluates the sample vector using neuro-linguistic analysis.
17. The method of claim 13 , wherein the values from each of the feature vectors include at least one of: a color feature, an entropy feature, a homogeneity feature, or a gradient feature.
18. The method of claim 13 , further comprising:
receiving, at a first time, a profile specifying a spatial granularity for at least one feature from the plurality of features of the first image; and
receiving, at a second time after the first time, the first image.
19. The method of claim 13 , further comprising:
receiving a profile specifying a spatial granularity for at least one feature from the plurality of features of the first image; and
adjusting a non-zero number of regions of the first image based on the spatial granularity.
20. The method of claim 13 , further comprising receiving the first image.