An intraoral light physical distance measurement method, device and readable storage medium

Abstract

The present application relates to the field of measurement, in particular to an intraoral radiography physical distance measurement method, device and readable storage medium, the present application is integrated with a ring surrounding the lens through the periphery of intraoral radiography, the ring has two or two and the mark point of different color with the ring. At the same time, with the aid of neural network, the depth map of intraoral radiography can be obtained, based on the depth map, the internal parameter of camera and the scale of shooting, the actual distance of two points in intraoral radiography can be obtained by calculation. Since the tooth is not in the same picture plane, the depth calculation makes the depth more fine, and the ruler does not need to be inserted into the patient's mouth to measure the actual distance.

Claims

1. A method for measuring physical distance via intraoral illumination, characterized in that... Includes the following steps: (1) Obtain a first intraoral photograph with an inner wall of an opening device. The opening device is equipped with a shooting device. The opening device is nested outside the camera of the shooting device. The shooting device is cylindrical with an elliptical cross-section. Several protruding marking points are provided on both sides of the inner wall of the opening device. Two of the marking points are selected as the two opening device marking points for the first intraoral photograph. (2) Input the first intraoral image into the trained target detection network to predict the position of the two mouthpiece markers in the first intraoral image and the planar image distance between the two mouthpiece markers that is perpendicular to the lens viewing direction. (3) Obtain a second intraorbital image with an inner wall of an opening device. The second intraorbital image has two second intraorbital image markers a1 and a2. Input the second intraorbital image into a trained depth estimation network to predict the depth map corresponding to the second intraorbital image and obtain the depth difference between two of the second intraorbital image markers. (4) First, convert the planar image distance between the two opening device markers in step (2) into the actual planar distance according to the scale. Then, by using the Pythagorean theorem, the depth difference between the two second in-orifice illumination markers obtained above, the position of the two opening device markers in the first in-orifice illumination, and the actual planar distance between the two opening device markers, the actual distance between the two second in-orifice illumination markers is obtained.

2. The method for measuring physical distance via intraoral illumination according to claim 1, characterized in that: The target detection network is trained, and the specific training process is as follows: (a) Take a first intraoral photograph of different users with the inner wall of the mouth opener through the mouth opener, the first intraoral photograph being a first intraoral photograph with at least two mouth opener markers, and marking the positions of the two mouth opener markers; (b) The first intraoral image with at least two mouth opening markers and the data of the positions of the two mouth opening markers marked thereon are fed into the target detection network to predict the positions of the two mouth opening markers; (c) The offset between the positions of the two marked opening points and the predicted positions of the two opening points is used as the loss function; wherein the loss function is used to guide the parameter update of the target detection network to obtain an accurate target detection network.

3. The method for measuring physical distance via intraoral illumination according to claim 2, characterized in that: The target detection network uses YOLOv5 as its network architecture.

4. The method for measuring physical distance via intraoral illumination according to claim 1, characterized in that: The depth estimation network is trained as follows: (a) Collect training data, each training data contains a second internal image with the inner wall of the opening device and the actual depth map corresponding to the second internal image; (b) Input the data of the second aperture internal image with the inner wall of the aperture and the depth map corresponding to the second aperture internal image into the depth estimation network to predict the corresponding depth image; (c) Input the predicted depth image and the actual corresponding depth map into the loss function calculation formula; wherein the loss function is used to guide the parameter update of the depth estimation network to obtain an accurate depth estimation network.

5. The method for measuring physical distance via intraoral illumination according to claim 4, characterized in that: The depth estimation network uses UNet as its network structure.

6. The method for measuring physical distance via intraoral illumination according to claim 1, characterized in that: In step (4), the calculation process for obtaining the actual distance between the two second-mouth inner-marker points is as follows: (a) Using the target detection network in step (2), the positions of the two opening markers in the first opening are obtained. The planar image distance between the two opening markers is calculated using the Pythagorean theorem and denoted as L1. The actual distance between the two opening markers is known and denoted as L2. The camera intrinsic parameters are known and denoted as K. (b) Input the second intraocular lens into the depth estimation network to obtain the predicted depth map, thereby obtaining the depths of the two second intraocular lens markers. Given the two second intraocular lens markers a1 and a2, obtain the depths of a1 and a2 in the map, denoted as D. a1 and D a2 Subtracting the two gives their depth difference, denoted as Dy; using the Pythagorean theorem, the pixel distance between the two second-internal illumination markers is calculated, denoted as L3; (c) The actual distance between the two second-aperture internal illumination markers on a plane perpendicular to the lens viewing direction is denoted as Dx. The depth Da1 of one of the second-aperture internal illumination markers is used as the input to the camera intrinsic parameter formula to eliminate lens distortion, thus obtaining K(D a1 The formula for calculating Dx is as follows: ； Finally, using the Pythagorean theorem, the actual distance between the two second-intercept markers a1 and a2 is calculated using Dy and Dx.

7. The method for measuring physical distance via intraoral illumination according to claim 1, characterized in that: The bottom of the opening device is provided with a buckle, and the bottom of the opening device is consistent with the size and shape of the shooting device. The bottom of the opening device is used to lock the shooting device.

8. An intraoral physical distance measuring device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be executed by the processor to implement an intraoral physical distance measurement method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The device contains a computer program that can be executed by a processor of the device in which the computer-readable storage medium is located, to implement the intraoral physical distance measurement method as described in any one of claims 1 to 7.

Images