Image processing device, image processing method, and program for image processing

An image processing method for correcting an amount of shake between pictures obtained by temporally-continuous capturing, the image processing device comprising:

• estimating a first shake amount indicating a rotational displacement amount of a second picture with respect to a first picture, the rotational displacement amount being obtained based on a combination of axis rotational directions of mutually perpendicular three axes, and the second picture being captured temporally after the first picture;
• setting an area to be used for post-processing for each of the first picture and a first corrected picture, using feature points extracted from the first picture and the second picture, the first corrected picture being generated from the second picture whose displacement is corrected using the estimated first shake amount;
• performing the post-processing, by generating a first cutout picture and a second cutout picture and estimating a second shake amount indicating a displacement amount of the second cutout picture with respect to the first cutout picture, the first cutout picture being generated from the area set in the first picture and the second cutout picture being generated from the area set in the first corrected picture;
• obtaining a shake amount indicating the amount of shake between the first picture and the second picture, from the estimated first shake amount and the estimated second shake amount;
• and
• correcting the amount of shake between the first picture and the second picture using the shake amount obtained by the obtaining of the shake amount,
• wherein the estimating includes:
• extracting the feature points from each of the first picture and the second picture;
• correcting distortion (i) developed due to optical strain of an optical system used for capturing and (ii) found between the feature points of the first picture and the feature points of the second picture, the feature points being extracted in the extracting;
• and
• matching the feature points of the first picture and the feature points of the second picture with each other to estimate the first shake amount, the first picture and the second picture having the distortion corrected in the correcting.

A non-transitory computer-readable recording medium having a program recorded thereon, the program for causing a computer to correct an amount of shake between pictures obtained by temporally-continuous capturing, the computer correcting the amount of shake by executing a method comprising:

• estimating a first shake amount indicating a rotational displacement amount of a second picture with respect to a first picture, the rotational displacement amount being obtained based on a combination of axis rotational directions of mutually perpendicular three axes, and the second picture being captured temporally after the first picture;
• setting an area to be used for post-processing for each of the first picture and a first corrected picture, using feature points extracted from the first picture and the second picture, the first corrected picture being generated from the second picture whose displacement is corrected using the estimated first shake amount;
• performing the post-processing, by generating a first cutout picture and a second cutout picture and estimating a second shake amount indicating a displacement amount of the second cutout picture with respect to the first cutout picture, the first cutout picture being generated from the area set in the first picture and the second cutout picture being generated from the area set in the first corrected picture;
• obtaining a shake amount indicating the amount of shake between the first picture and the second picture, from the estimated first shake amount and the estimated second shake amount;
• and
• correcting the amount of shake between the first picture and the second picture using the shake amount obtained by obtaining of the shake amount,
• wherein the estimating includes:
• extracting the feature points from each of the first picture and the second picture;
• correcting distortion (i) developed due to optical strain of an optical system used for capturing and (ii) found between the feature points of the first picture and the feature points of the second picture, the feature points being extracted in the extracting;
• and
• matching the feature points of the first picture and the feature points of the second picture with each other to estimate the first shake amount, the first picture and the second picture having the distortion corrected in the correcting.