Dynamic image selection

摘要

Image capture can be improved by capturing a sequence of images and analyzing the images to select the image with the least blur and/or an acceptable amount of blur. Gradients can be calculated for at least a portion of the images, and gradient histograms generated. Two or more component curves can be fit to each histogram, such as by using a Gaussian mixture model, and the curves can be compared to determine an amount of variation between the curves. The image with the smallest differences between component curves, or with differences less than a specified blur threshold, can be selected as a sufficiently sharp image and provided for viewing, processing, or another intended purpose of the image to be captured.

主权项

1. A computer-implemented method of selecting an image, comprising:
receiving an instruction to capture an image using a camera of a computing device; causing a sequence of images to be captured in response to the command; for each image in the sequence of images:
calculating a set of image gradients for the image;generating a gradient histogram using the set of image gradients; andidentifying a first frequency range indicative of sharp features in the image and a second frequency range indicative of blurry features in the imagefitting a first curve and a second curve to the gradient histogram using a two-component Gaussian mixture model, the first curve corresponding to the first specified frequency range and the second curve corresponding to the second specified frequency range; and selecting an image of the plurality of images to provide for processing, the selected image having a smallest difference between the first curve and the second curve from among the images in the sequence of images.