Method, apparatus, and recording medium for evaluating reference points, and method, apparatus, and recording medium for positional alignment
AssigneeFUJIFILM Corporation
InventorTsutomu Inoue
Life of this patent
Estimating the term…
Abstract
An initial position aligning unit performs initial positional alignment of a video and simulation data, and a position aligning unit performs positional alignment of the video and the simulation data. A first movement detecting unit detects movements of reference points which are set within a first image of the video, and a second movement detecting unit detects movement of a camera. A degree of reliability calculating unit compares the first and second movements to calculate a degree of reliability with respect to the movements of each of the reference points.
Claims19
A position aligning apparatus, comprising:
- a camera for obtaining a video constituted by at least a first image and a second image, the first image and the second image including an image of a structure within a human body and having different imaging times;
- and
- at least one processor configured to:
- compare the first image and the second image of a different time than the first image, from among images that constitute the video, and to detect the movements between each of a plurality of reference points within the first image and corresponding reference points included in the second image as first movements, the movements occurring between a time that the first image is obtained and a time that the second image is obtained;
- detect movements between one or more invariable points within the first image and corresponding invariable points included in the second image as second movements, which is the movement of the camera, the movements occurring between the time that the first image is obtained and the time that the second image is obtained, wherein the invariable points are within the imaging range of the camera that do not change as imaging of the video progresses, and wherein the invariable points are different from the reference points;
- compare the first movements and the second movement, and for calculating a degree of reliability with respect to the movements of each of the reference points;
- obtain simulation data of target object included in the video;
- perform initial positional alignment of the target object included in the video and the simulation data;
- perform positional alignment of the target object included in the video and the simulation data based on the second movement;
- and
- display aligned simulation data and the video on which the simulation data is overlapped,
- wherein the at least one processor performs a first judgment regarding whether it is possible to detect the second movement, performs positional alignment of the target object included in the video and the simulation data based on the second movement in the case that the result of the first judgment is affirmative, and performs positional alignment of the target object included in the video and the simulation data based on first movements of reference points having degrees of reliability greater than or equal to a second threshold value, which is set in advance, in the case that the result of the first judgment is negative.
A reference point evaluating method comprising:
- obtaining a video constituted by at least a first image and a second image, the first image and the second image including an image of a structure within a human body and having different imaging times;
- comparing the first image and the second image of a different time than the first image, from among images that constitute the video, and detecting the movements between each of a plurality of reference points within the first image and corresponding reference points included in the second image as first movements, the movements occurring between a time that the first image is obtained and a time that the second image is obtained;
- detecting movements between one or more invariable points within the first image and corresponding invariable points included in the second image as second movements, which is the movement of the camera, the movements occurring between the time that the first image is obtained and the time that the second image is obtained, wherein the invariable points are within the imaging range of the camera that do not change as imaging of the video progresses, and wherein the invariable points are different from the reference points;
- comparing the first movements and the second movement, and calculating a degree of reliability with respect to the movements of each of the reference points;
- obtaining simulation data of target object included in the video;
- performing initial positional alignment of the target object included in the video and the simulation data;
- performing positional alignment of the target object included in the video and the simulation data based on the second movement;
- and
- displaying aligned simulation data and the video on which the simulation data is overlapped,
- wherein the at least one processor performs a first judgment regarding whether it is possible to detect the second movement, performs positional alignment of the target object included in the video and the simulation data based on the second movement in the case that the result of the first judgment is affirmative, and performs positional alignment of the target object included in the video and the simulation data based on first movements of reference points having degrees of reliability greater than or equal to a second threshold value, which is set in advance, in the case that the result of the first judgment is negative.
A non-transitory recording medium having a reference point evaluating program recorded therein, the program causing a computer to execute the procedures of:
- obtaining a video constituted by at least a first image and a second image, the first image and the second image including an image of a structure within a human body having different imaging times;
- comparing the first image and the second image of a different time than the first image, from among images that constitute the video, and detecting the movements between each of a plurality of reference points within the first image and corresponding reference points included in the second image as first movements, the movements occurring between a time that the first image is obtained and a time that the second image is obtained;
- detecting movements between one or more invariable points within the first image and corresponding invariable points included in the second image as second movements, which is the movement of the camera, the movements occurring between the time that the first image is obtained and the time that the second image is obtained, wherein the invariable points are within the imaging range of the camera that do not change as imaging of the video progresses, and wherein the invariable points are different from the reference points;
- comparing the first movements and the second movement, and calculating a degree of reliability with respect to the movements of each of the reference points;
- obtaining simulation data of target object included in the video;
- performing initial positional alignment of the target object included in the video and the simulation data;
- performing positional alignment of the target object included in the video and the simulation data based on the second movement;
- and
- displaying aligned simulation data and the video on which the simulation data is overlapped,
- wherein the at least one processor performs a first judgment regarding whether it is possible to detect the second movement, performs positional alignment of the target object included in the video and the simulation data based on the second movement in the case that the result of the first judgment is affirmative, and performs positional alignment of the target object included in the video and the simulation data based on first movements of reference points having degrees of reliability greater than or equal to a second threshold value, which is set in advance, in the case that the result of the first judgment is negative.
A reference point evaluating apparatus, comprising:
- a camera for obtaining a video constituted by at least a first image and a second image, the first image and the second image including an image of a structure within a human body having different imaging times;
- and
- at least one processor configured to:
- compare a first image and a second image of a different time than the first image, from among images that constitute the video, and detect the movements of each of a plurality of reference points within the first image between the first image and the second image as first movements;
- obtain simulation data of target object included in the video;
- perform initial positional alignment of the target object included in the video and the simulation data;
- perform positional alignment of the target object included in the video and the simulation data based on the second movement;
- and
- display aligned simulation data and the video on which the simulation data is overlapped,
- wherein the at least one processor performs a first judgment regarding whether it is possible to detect the second movement, performs positional alignment of the target object included in the video and the simulation data based on the second movement in the case that the result of the first judgment is affirmative, and performs positional alignment of the target object included in the video and the simulation data based on first movements of reference points having degrees of reliability greater than or equal to a second threshold value, which is set in advance, in the case that the result of the first judgment is negative, and
- a motion sensor that detects movement of the camera between a time when the first image was obtained and a time when the second image was obtained as a second movement;
- wherein the processor is further configured to compare the first movements and the second movement, and calculate a degree of reliability with respect to the movements of each of the reference points.
A reference point evaluating method, comprising:
- obtaining a video, captured via a camera, constituted by at least a first image and a second image, the first image and the second image including an image of a structure within a human body having different imaging times;
- and
- compare, via at least one processor, a first image and a second image of a different time than the first image, from among images that constitute the video, and detect the movements of each of a plurality of reference points within the first image between the first image and the second image as first movements;
- detect movement of the camera, via a motion sensor that detects movement of the camera, between a time when the first image was obtained and a time when the second image was obtained as a second movement;
- compare, via the at least one processor, the first movements and the second movement, and calculate a degree of reliability with respect to the movements of each of the reference points;
- obtain simulation data of target object included in the video;
- perform initial positional alignment of the target object included in the video and the simulation data;
- perform positional alignment of the target object included in the video and the simulation data based on the second movement;
- and
- display aligned simulation data and the video on which the simulation data is overlapped,
- wherein the at least one processor performs a first judgment regarding whether it is possible to detect the second movement, performs positional alignment of the target object included in the video and the simulation data based on the second movement in the case that the result of the first judgment is affirmative, and performs positional alignment of the target object included in the video and the simulation data based on first movements of reference points having degrees of reliability greater than or equal to a second threshold value, which is set in advance, in the case that the result of the first judgment is negative.
A non-transitory recording medium having a position aligning program recorded therein, the program causing a computer to execute the procedures of:
- obtaining a video, captured via a camera, constituted by at least a first image and a second image, the first image and the second image including an image of a structure within a human body having different imaging times;
- and
- compare, via at least one processor, a first image and a second image of a different time than the first image, from among images that constitute the video, and detect the movements of each of a plurality of reference points within the first image between the first image and the second image as first movements;
- detect movement of the camera, via a motion sensor that detects movement of the camera, between a time when the first image was obtained and a time when the second image was obtained as a second movement;
- compare, via the at least one processor, the first movements and the second movement, and calculate a degree of reliability with respect to the movements of each of the reference points;
- obtain simulation data of target object included in the video;
- perform initial positional alignment of the target object included in the video and the simulation data;
- perform positional alignment of the target object included in the video and the simulation data based on the second movement;
- and
- display aligned simulation data and the video on which the simulation data is overlapped,
- wherein the at least one processor performs a first judgment regarding whether it is possible to detect the second movement, performs positional alignment of the target object included in the video and the simulation data based on the second movement in the case that the result of the first judgment is affirmative, and performs positional alignment of the target object included in the video and the simulation data based on first movements of reference points having degrees of reliability greater than or equal to a second threshold value, which is set in advance, in the case that the result of the first judgment is negative.
Dependencies are inferred from the claim text.