Foot feature point identification method based on three-dimensional point cloud

Abstract

The invention discloses a foot feature point identification method based on a three-dimensional point cloud. The method comprises the following steps: extracting a principal axis in a projected sole point cloud based on a foot three-dimensional point cloud; connecting lines of the first datum point, the second datum point and the third datum point form a datum plane; projecting the main shaft to a reference plane to obtain a foot axis, and establishing a reference coordinate system based on the foot axis and the reference plane; and determining feature points in the foot three-dimensional point cloud based on the reference coordinate system and foot parameters determined based on the feature points. According to the foot feature point identification method, the reference coordinate system is established based on the reference plane determined by the foot axis and the three extreme points, foot shape parameters with space curved surface foot sole features can be accurately measured, and foot shape parameter measurement is not affected by ankle joint rotation. The principal component analysis (PCA) method is adopted to determine the principal axis, and then the foot axis is determined through the projection of the principal axis on the reference plane, so that the foot axis is not affected by metatarsophalangeal deformity and the fat and thin degree.

Description

Technical Field

[0001] The present invention belongs to the field of foot measurement and relates to a foot feature point recognition method based on three-dimensional point cloud. Background Art

[0002] The identification and accurate measurement of foot morphology parameters are fundamental to the clinical diagnosis of foot disorders, the adaptation of intelligent assistive devices, and the design of personalized assistive devices. Traditional foot morphology measurement methods use calipers, tape measures, and footprints. While simple, these methods suffer from limited accuracy and efficiency. Three-dimensional digital scanning technology has made foot morphology measurement more precise and efficient. This technology can rapidly acquire three-dimensional foot data, providing parameters in multiple dimensions such as foot length, width, and circumference, while reducing errors caused by human contact. 3D scanning technologies vary, including stereo photogrammetry based on multi-angle photographs, structured light technology that measures by projecting specific patterns, laser scanning technology that uses continuous laser projection in conjunction with cameras and computers to capture three-dimensional information, and more recently, depth camera solutions that can continuously measure dynamic morphology, even using a mobile phone. Three-dimensional scanning of foot morphology typically requires a stationary standing position, with the sole of the foot interacting with a rigid surface rather than a flexible interface like an insole. Consequently, the resulting morphology does not reflect the actual deformation of the foot within the shoe. This scanning method leads to inaccurate foot morphology during the detection phase, and using the scanner's acquisition plane as the reference plane for identifying and measuring foot feature points also reduces the accuracy of foot morphology parameter measurements. Non-weight-bearing methods such as suspended profiling can obtain the plantar surface morphology, but this method does not produce contact and the sole of the foot does not deform. The non-deformation of the plantar soft tissue or the deformation on a hard plate will lead to inaccurate foot morphology parameter measurements. Existing foot feature point recognition methods usually use the acquisition plane as the reference plane for identifying and measuring foot feature points, which makes it difficult to accurately identify foot models with weight-bearing deformed plantar surfaces. Summary of the Invention

[0003] In order to overcome the deficiencies of the prior art, the present invention provides a foot feature point recognition method based on three-dimensional point cloud.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for identifying foot feature points based on a three-dimensional point cloud comprises the following steps:

[0006] Step 1: Based on the 3D point cloud of the foot, select within a set height range to obtain the plantar point cloud. Project the plantar point cloud onto a 2D plane perpendicular to the direction of gravity, and extract the principal axis of the projected plantar point cloud using principal component analysis.

[0007] Step 2: Divide the plantar point cloud into the forefoot region and the hindfoot region using the perpendicular bisector of the principal axis as the boundary. Set the point with the minimum height in the gravity direction of the hindfoot region as the first reference point. Set the points with the minimum height in the gravity direction of the forefoot region on both sides of the principal axis as the second and third reference points. Connect the first, second, and third reference points to form a reference plane.

[0008] Step 3: Project the principal axis onto the reference plane to obtain the foot axis, and establish a reference coordinate system based on the foot axis and the reference plane;

[0009] Step 4: Determine the feature points in the three-dimensional point cloud of the foot based on the reference coordinate system and the foot parameters determined based on the feature points.

[0010] Furthermore, in step 3, the origin of the reference coordinate system can be selected as any point on the foot axis.

[0011] Furthermore, in step 4, the feature points include global extreme points, local extreme points, and arch extreme points.

[0012] Furthermore, the extreme points of the arch of the foot include the highest point of the medial longitudinal arch and the highest point of the lateral longitudinal arch. The method for determining the extreme points of the arch of the foot includes: determining the first metatarsal point, the fifth metatarsal point, and the heel point based on the reference coordinate system to determine the medial longitudinal arch plane and the lateral longitudinal arch plane; wherein, the medial longitudinal arch plane is defined as passing through the heel point and the first metatarsal point and perpendicular to the reference plane, and the point where the medial longitudinal arch plane intersects with the plantar curved surface has the maximum height as the highest point of the medial longitudinal arch; the lateral longitudinal arch plane is defined as passing through the heel point and the fifth metatarsal point and perpendicular to the reference plane, and the point where the lateral longitudinal arch plane intersects with the plantar curved surface has the maximum height as the highest point of the lateral longitudinal arch.

[0013] Furthermore, the global extreme point is obtained by selecting the maximum value corresponding to the reference coordinate system direction of the three-dimensional point cloud of the foot.

[0014] Furthermore, the local extreme points are first selected in the corresponding local area based on the foot shape patterns of adult men and women across the country, and then the extreme points in a specific direction in the reference coordinate system are found in the local area.

[0015] In summary, the present invention is beneficial in that:

[0016] The foot feature point recognition of the present invention establishes a reference coordinate system based on the reference plane determined by the foot axis and three extreme points, which can accurately measure the foot morphological parameters with spatial curved surface plantar features, and the foot morphological parameter measurement is not affected by ankle joint rotation.

[0017] The present invention adopts a principal component analysis method to determine the principal axis, and then determines the foot axis by projecting the principal axis on a reference plane, so that the foot axis is not affected by metatarsophalangeal deformity and fatness or thinness.

[0018] The definition and identification method of the arch of the foot of the present invention is based on bony characteristics and is determined according to the extreme points of the calcaneal and phalangeal regions. It accurately matches the definition of arch height in the medical field and is more in line with foot anatomy and physiology and clinical diagnostic standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of setting the reference plane of the present invention.

[0020] Figure 2 Schematic diagram of the bottom view of the method for determining the extreme point of the arch of the foot according to the present invention.

[0021] Figure 3 It is a front view schematic diagram of the method for determining the extreme point of the arch of the foot of the present invention. DETAILED DESCRIPTION

[0022] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0023] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be arbitrarily changed, and the component layout may also be more complex.

[0024] All directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, horizontal, vertical...) are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0025] The present invention provides a method for identifying foot feature points based on a three-dimensional point cloud, comprising the following steps:

[0026] Step 1: Based on the 3D point cloud of the foot, select within a set height range to obtain the plantar point cloud. Project the plantar point cloud onto a 2D plane perpendicular to the direction of gravity, and extract the principal axis of the projected plantar point cloud using principal component analysis.

[0027] The three-dimensional point cloud of the foot can be obtained through 3D scanning technology, such as using a UPOD laser scanner, and the scanned point cloud is processed into a three-dimensional point cloud of the foot through the PCL (Point Cloud Library) library. The above method for obtaining the three-dimensional point cloud of the foot is a direct reference to the existing technology and will not be described in detail.

[0028] Specifically, for the three-dimensional point cloud of the foot, first select the approximate range of the plantar point cloud, then project the point cloud onto a two-dimensional plane perpendicular to the direction of gravity, and then use PCA principal component analysis to extract the principal axis of the projected point cloud.

[0029] Step 2: Divide the plantar point cloud into the forefoot area and the hindfoot area with the perpendicular median of the main axis as the boundary. Set the minimum height point in the gravity direction of the hindfoot area as the first reference point, and the minimum height points in the gravity direction of the forefoot area on both sides of the main axis as the second and third reference points. The line connecting the first, second and third reference points forms a reference plane. Figure 1 shown.

[0030] In order to reduce the reference plane error, the foot is divided into two areas, front and back, and reference points are selected in each area for measurement to avoid errors caused by excessive concentration of reference points.

[0031] Given that the body's weight is primarily borne by the calcaneus and metatarsal regions, by distinguishing between the forefoot and hindfoot, the lowest support points located in the calcaneus and metatarsal regions can be accurately identified.

[0032] Similarly, by dividing the forefoot along the main axis, the lowest support points on each side of the metatarsal bones can be determined, and these three points together define the reference plane.

[0033] The extreme points of gravity in the area are selected as reference points to ensure that these three points are the lowest points supporting the body weight. This is done to ensure that all point cloud data is above the reference plane, thus effectively avoiding the occurrence of negative distances.

[0034] Different from the traditional hard-board plane measurement, the measurement of the foot in the weight-bearing position under the flexible interface may cause the ankle joint to rotate, so the plantar plane is spatially changing. The traditional method is based on a fixed plantar plane to identify and measure the foot feature points, which cannot be applied to the plantar plane that changes in space. The method of determining the reference plane of the present invention can determine a reference plane that is always parallel to the sole of the foot. The identification and measurement of the foot feature points are performed based on the reference plane. For example, the measurement method of the feature points related to height is the distance from the point to the reference plane; the measurement method of the feature points related to length is the distance between points in the direction of the coordinate axis; the measurement method of the feature points related to angle is the angle between the line between points and the direction of the coordinate axis.

[0035] Step 3: Project the principal axis onto the reference plane to obtain the foot axis. Establish a reference coordinate system based on the foot axis and the reference plane. The origin of the reference coordinate system can be any point on the foot axis. Projecting the principal axis onto the reference plane to obtain the foot axis ensures that the foot axis remains parallel to the foot and is unaffected by ankle rotation.

[0036] Step 4: Determine the feature points in the three-dimensional point cloud of the foot and the foot parameters determined based on the feature points based on the reference coordinate system, wherein the feature points include but are not limited to global extreme points, local extreme points, and arch extreme points.

[0037] Among them, the global extreme point is obtained by selecting the maximum value of the three-dimensional point cloud of the foot corresponding to the direction of the reference coordinate system; the local extreme point is first selected from the corresponding local area based on the foot shape rules of adult men and women across the country, and then the extreme point in a specific direction is found in the local area.

[0038] Furthermore, in the present invention, the arch extreme points include the highest point of the medial longitudinal arch and the highest point of the lateral longitudinal arch. To determine the arch extreme points, first determine the first metatarsal point, the fifth metatarsal point, and the heel point based on the reference coordinate system, and then refer to Figure 2 and Figure 3 As shown, where:

[0039] Determine the medial longitudinal arch plane, which is defined as the plane passing through the heel point and the first metatarsal point and perpendicular to the reference plane. The point where this plane intersects the plantar surface with the maximum height is the highest point of the medial longitudinal arch. The distance from the highest point of the medial longitudinal arch to the reference plane is the medial longitudinal arch height.

[0040] Determine the lateral longitudinal arch plane. The lateral longitudinal arch plane is defined as the point passing through the heel point and the fifth metatarsal point and perpendicular to the reference plane. The point where the plane intersects the plantar surface with the maximum height is the highest point of the lateral longitudinal arch. The distance from the highest point of the lateral longitudinal arch to the reference plane is the lateral longitudinal arch height.

[0041] The present invention determines the extreme points of the arch by first locating skeletal feature points, then accurately calculating the vertical distance from the plantar support plane to the arch apex, thereby determining the absolute height of the arch. This method precisely aligns with the medical definition of arch height, which measures the actual distance from the support plane formed by the heel and forefoot metatarsal heads to the arch apex.

[0042] Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

Claims

1. A foot feature point recognition method based on three-dimensional point cloud, characterized in that: The following steps are involved: Step 1: Based on the 3D point cloud of the foot, select within a set height range to obtain the plantar point cloud. Project the plantar point cloud onto a 2D plane perpendicular to the direction of gravity, and extract the principal axis of the projected plantar point cloud using principal component analysis. Step 2: Divide the plantar point cloud into the forefoot region and the hindfoot region using the perpendicular bisector of the principal axis as the boundary. Set the point with the minimum height in the gravity direction of the hindfoot region as the first reference point. Set the points with the minimum height in the gravity direction of the forefoot region on both sides of the principal axis as the second and third reference points. Connect the first, second, and third reference points to form a reference plane. Step 3: Project the principal axis onto the reference plane to obtain the foot axis, and establish a reference coordinate system based on the foot axis and the reference plane; Step 4: Determine the feature points in the three-dimensional point cloud of the foot based on the reference coordinate system and the foot parameters determined based on the feature points.

2. The method for identifying foot feature points based on three-dimensional point cloud according to claim 1, characterized in that: In step 3, the origin of the reference coordinate system can be selected as any point on the foot axis.

3. The foot feature point recognition method based on three-dimensional point cloud according to claim 1, characterized in that: In step 4, the feature points include global extreme points, local extreme points, and arch extreme points.

4. The method for identifying foot feature points based on three-dimensional point cloud according to claim 3, characterized in that: The extreme points of the arch of the foot include the highest point of the medial longitudinal arch and the highest point of the lateral longitudinal arch. The method for determining the extreme points of the arch of the foot includes: determining the first metatarsal point, the fifth metatarsal point, and the heel point based on the reference coordinate system to determine the medial longitudinal arch plane and the lateral longitudinal arch plane; wherein, the medial longitudinal arch plane is defined as passing through the heel point and the first metatarsal point and perpendicular to the reference plane, and the point where the medial longitudinal arch plane intersects with the plantar curved surface has the maximum height as the highest point of the medial longitudinal arch; the lateral longitudinal arch plane is defined as passing through the heel point and the fifth metatarsal point and perpendicular to the reference plane, and the point where the lateral longitudinal arch plane intersects with the plantar curved surface has the maximum height as the highest point of the lateral longitudinal arch.

5. The method for identifying foot feature points based on three-dimensional point cloud according to claim 3, characterized in that: The global extreme point is obtained by selecting the maximum value corresponding to the reference coordinate system direction of the three-dimensional point cloud of the foot.

6. The method for identifying foot feature points based on three-dimensional point cloud according to claim 3, characterized in that: The local extreme points are first selected based on the foot shape patterns of adult men and women across the country in the corresponding local area, and then the extreme points in a specific direction in the reference coordinate system are found in the local area.

Images