Response normalization for overlapped multi-image applications
Systems and methods for registration of image pairs, including camera response function normalization are disclosed and include selecting a pair of images from a set of images, each of the pair of images having associated metadata, determining a camera response function for each of the images in the pair of images using the associated metadata, normalizing each camera response function for across the set of images, and applying the normalized camera response function to the pair of images. A deformation map is generated in a multi-scale process using registration parameters, to deform one of the pair of images to align with another of the pair of images. The image pairs may be selected by identifying image pairs having an overlap that exceeds an overlap threshold, having a sequential proximity in an image series satisfying a proximity threshold, and/or having estimated image capture attitudes within an attitude threshold.
1. A method comprising:
identifying a plurality of image pairs from a set of images, each image pair comprising a first image selected from the set of images and a second image selected from the set of images that satisfies image registration criteria with the first image;
determining, separately for each of the image pairs, a normalized camera response function for each of the images in the image pair;
generating an optimized normalized camera response function for each image of the set of images by optimizing the normalized camera response functions determined for two or more image pairs that include the image; and
applying, for each image in the set of images, the optimized normalized camera response function.
2. The method of claim 1 , wherein the plurality of image pairs are infrared images.
3. The method of claim 1 , further comprising normalizing a dynamic range of each of the plurality of image pairs.
4. The method of claim 1 , further comprising applying at least one lens distortion parameter corresponding to components of an image capture device to each image of the set of images.
5. The method of claim 1 , further comprising demixing each of the plurality of image pairs to remove low-frequency non-uniformity.
6. The method of claim 1 , further comprising creating a deformation map using registration parameters, wherein the deformation map is used to align the second image with the first image.
7. The method of claim 6 , wherein creating the deformation map further comprises a multi-scale process comprising resizing the images to a current processing scale of the multi-scale process.
8. The method of claim 7 , wherein the multi-scale process further comprises, after resizing images to a current processing scale of the multi-scale process, optimizing parameters for the resized images, applying the camera response function, and deforming the resized images.
9. The method of claim 8 , wherein the multi-scale process further comprises updating the deformation map in accordance with an error histogram.
10. The method of claim 1 , wherein identifying the plurality of image pairs from the set of images comprises:
identifying image pairs having an overlap that exceeds an overlap threshold, having a sequential proximity in an image series satisfying a proximity threshold, and/or having estimated image capture attitudes within an attitude threshold.
11. A system comprising:
a processor; and
a local memory operable to store a plurality of machine-readable instructions which when executed by the processor are operable to cause the system to perform steps comprising:
identifying a plurality of image pairs from a set of images, each image pair comprising a first image selected from the set of images and a second image selected from the set of images that satisfies image registration criteria with the first image;
determining, separately for each of the image pairs, a normalized camera response function for each of the images in the image pair;
generating an optimized normalized camera response function for each image of the set of images by optimizing the normalized camera response functions determined for two or more image pairs that include the image; and
applying, for each image in the set of images, the optimized normalized camera response function.
12. The system of claim 11 , wherein the plurality of image pairs are infrared images.
13. The system of claim 11 , wherein the plurality of machine-readable instructions, when executed by the processor, are further operable to cause the system to perform steps comprising:
normalizing a dynamic range of each of the plurality of image pairs.
14. The system of claim 11 , wherein the plurality of machine-readable instructions, when executed by the processor, are further operable to cause the system to perform steps comprising:
applying at least one lens distortion parameter corresponding to components of an image capture device to each image of the set of images.
15. The system of claim 11 , wherein the plurality of machine-readable instructions, when executed by the processor, are further operable to cause the system to perform steps comprising:
demixing each of the plurality of image pairs to remove low-frequency non-uniformity.
16. The system of claim 11 , wherein the plurality of machine-readable instructions, when executed by the processor, are further operable to cause the system to perform steps comprising:
creating a deformation map using registration parameters, wherein the deformation map is used to align the second image with the first image.
17. The system of claim 16 , wherein creating the deformation map further comprises a multi-scale process comprising resizing the images to a current processing scale of the multi-scale process.
18. The system of claim 17 , wherein the multi-scale process further comprises, after resizing the images to the current processing scale of the multi-scale process, optimizing parameters for the resized images, applying the corresponding camera response function, and deforming the resized images.
19. The system of claim 18 , wherein the multi-scale process further comprises updating the deformation map in accordance with an error histogram.
20. The system of claim 11 , wherein selecting the plurality of image pairs from the set of images comprises identifying image pairs having an overlap that exceeds an overlap threshold, having a sequential proximity in an image series satisfying a proximity threshold, and/or having estimated image capture attitudes within an attitude threshold.