TWI772726B - Assistive-device modeling method and limb guide-plate mechamism - Google Patents

Abstract

An assistive-device modeling method includes the following steps. Point cloud information of a limb surface corresponding to a limb is obtained. A plurality of learning data are generated according to the point cloud information of the limb surface. The learning data are processed through a neural network to generate an assistive-device digital model.

Description

Assistive device modeling method and limb guide mechanism

The present invention relates to a modeling method, in particular, to a modeling method of an auxiliary device and a limb guide mechanism.

When a person's limbs are injured, their movements will be affected. At this time, some equipment (such as casts, crutches, protective gear, prosthetic limbs, etc.) is usually used for assistance, which is called an assistive device. Based on functional distinctions, assistive devices may include mobility aids (eg, wheelchairs, orthopedic shoes, walking aids, crutches), assistive devices that assist bodily functions (eg, hand or leg prostheses, pen holders, special spoons) ), life aids (such as safety alarm bells, kickers), therapeutic or fixed aids (such as plaster, back frame, waist support, knee support, collar, pillow)...etc. The more common and frequently used protective devices are usually assistive devices related to physical activities, such as limb protective devices, walking aids, prosthetics, casts, and the like.

Generally speaking, the aids for immobilizing the limbs are usually made by experienced doctors or medical personnel with plaster, mold or related objects, but such plaster-made braces will lead to a very poor user experience. For example, because the plaster hardens after cooling, the user's skin will be very uncomfortable when it comes into contact with the plaster; high temperature is generated when making plaster-based protective gear; the material or structure cannot be changed to achieve customization...etc. On the other hand, auxiliary devices can also be produced in a fixed-size method, but such auxiliary devices cannot provide perfect support, flexibility and comfort for the limbs in use. In addition, the material and structure of the uniformly produced auxiliary equipment cannot be adjusted or changed according to the needs of users.

Therefore, how to quickly build a digital model of the wearable assistive device, improve the design quality of the assistive device, and shorten the production cycle of the customized assistive device will become an important issue.

The present invention provides an auxiliary device modeling method and a limb guide mechanism, so as to quickly establish a digital model of the auxiliary device of the wearable auxiliary device, improve the design quality of the auxiliary device and shorten the production cycle of the customized auxiliary device.

The present invention provides a method for modeling an auxiliary device, which includes the following steps. Obtain the point cloud information of the limb surface of the corresponding limb. According to the point cloud information of the body surface, multiple learning materials are generated. Through a neural network, the learning data is processed to generate a digital model of the assistive device.

The present invention further provides a method for modeling an auxiliary device, which includes the following steps. A limb guide mechanism is provided, wherein the limb guide mechanism is provided with at least three positioning holes. At least three positioning marks corresponding to at least three positioning holes are generated on the limb through the limb guide mechanism. Scan the limb to obtain the point cloud information of the limb surface, and the point cloud information of the limb surface includes at least three positioning marks. Obtain multiple preset assistive digital models from the database. According to the at least three positioning marks of the point cloud information of the limb surface, the point cloud information of the limb surface and the digital model of the preset assistive device are superimposed. Select the digital model of the preset assistive device that matches the point cloud information of the limb surface. According to the point cloud information of the limb surface and the matching preset digital model of the assistive device, the digital model of the assistive device is generated.

The present invention further provides a limb guide mechanism, which includes a body and at least three positioning holes. The positioning holes are arranged on the body, and the positions of the positioning holes are different from each other, and the length of the connecting line between the two positioning holes of the at least three positioning holes is set to be less than or equal to a preset distance.

The assistive device modeling method and the limb guide mechanism disclosed in the present invention generate a plurality of learning data according to the point cloud information of the limb surface, and process the learning data through a neural network to generate a digital model of the assistive device. In addition, the embodiments of the present invention can provide a limb guide mechanism with at least three positioning holes, and generate at least three positioning marks corresponding to the at least three positioning holes on the limb through the limb guide mechanism. Next, scan the limb to obtain the point cloud information of the limb surface including at least three positioning marks, and according to the at least three positioning marks of the point cloud information of the limb surface, compare the point cloud information of the limb surface with a plurality of preset points obtained in the database. The assistive device digital model is set for superimposition, the preset assistive device digital model that matches the point cloud information of the limb surface is selected, and then the assistive device digital model is generated according to the limb surface point cloud information and the matching preset assistive device digital model. In this way, an auxiliary device digital model of the wearable auxiliary device can be established effectively and quickly, the design quality of the auxiliary device can be improved, and the production cycle of the customized auxiliary device can be shortened.

For the technical terms in this specification, reference is made to the common terms in the technical field. If some terms are described or defined in this specification, the interpretation of this part of the terms shall be subject to the descriptions or definitions in this specification. Each embodiment of the present disclosure has one or more technical features. Under the premise of possible implementation, those skilled in the art can selectively implement some or all of the technical features in any embodiment, or selectively combine some or all of the technical features in these embodiments.

In the various embodiments listed below, the same or similar elements or components will be represented by the same reference numerals.

FIG. 1 is a schematic diagram of an electronic system for modeling an assistive device according to an embodiment of the present invention. Please refer to FIG. 1 , an electronic system 100 for modeling auxiliary devices includes a 3D scanning device 110 , a processing device 120 and a 3D printing device 130 .

The three-dimensional scanning device 110 is used to scan the limb of the user to generate point cloud information of the limb surface corresponding to the limb. For example, the user can place the hand or the foot in the scanning area of the three-dimensional scanning device 110 for scanning, so as to generate the point cloud information of the limb surface corresponding to the limb (ie, the user's hand or foot).

The processing device 120 is coupled to the three-dimensional scanning device 110 to receive the point cloud information of the limb surface generated by the three-dimensional scanning device 110, and generate a suitable digital model of the assistive device according to the point cloud information of the limb surface.

In this embodiment, the processing device 120 may include a processor and a database. The processor is, for example, a central processing unit (Central Processing Unit, CPU), a microprocessor (Microprocessor), a microcontroller, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) ) or other similar device. The database is, for example, a memory, a hard disk, a solid-state hard disk, etc., for storing relevant information.

The three-dimensional printing device 130 is coupled to the processing device 120 and generates an assistive device suitable for the user according to the digital model of the assistive device generated by the processing device 120 . In this way, the electronic system 100 for modeling the assistive device can quickly create a digital model of the assistive device of the wearable assistive device, improve the design quality of the assistive device and shorten the production cycle of the customized assistive device.

FIG. 2 is a flowchart of an assistive device modeling method according to an embodiment of the present invention. Please refer to Figure 1 and Figure 2 at the same time. In step S202 , the processing device 120 obtains the point cloud information of the limb surface of the corresponding limb, as shown in the reference numeral 310 in FIG. 3 . The above body surface point cloud information is generated by scanning the user's body through the 3D scanning device 110 . In this embodiment, the user's hand is used as an example of the limb of the embodiment of the present invention.

In step S204, the processing device 120 generates a plurality of learning data according to the point cloud information of the body surface. Further, the processing device 120, for example, rotates the point cloud information of the body surface to a plurality of different angles and performs proportional scaling, so as to generate a plurality of learning data in sequence. For example, the processing device 120 may rotate the body surface point cloud information to a plurality of different angles through the rotation direction 310 shown in FIG. 3 to generate a plurality of learning data corresponding to the original scale. Next, the processing device 120 can perform proportional scaling of the point cloud information on the body surface, and rotate the above proportionally scaled body surface point cloud information to a plurality of different angles through the rotation direction 310 shown in FIG. 3 to generate Multiple learning materials corresponding to different scales and different angles.

In step S206, the processing device 120 processes the learning data through a neural network to generate a digital model of the assistive device. That is to say, the processing device 120 processes the above-mentioned learning data through, for example, a convolutional neural network to generate a corresponding limb model, and draws a digital model of the assistive device corresponding to the limb model according to the limb model, so as to generate a corresponding limb model. Assistive digital model.

In step S208, an auxiliary device is generated according to the digital model of the auxiliary device through the three-dimensional printing device 130. In this way, a digital model of the wearable assistive device can be quickly established, the design quality of the assistive device can be improved, and the production cycle of the customized assistive device can be shortened.

FIG. 4 is a schematic diagram of an electronic system for modeling an assistive device according to an embodiment of the present invention. Please refer to FIG. 4 , an electronic system 400 for modeling an assistive device includes a 3D scanning device 110 , a processing device 120 , a 3D printing device 130 and a limb guide mechanism 410 . In this embodiment, the 3D scanning device 110 , the processing device 120 , and the 3D printing device 130 are the same as or similar to the 3D scanning device 110 , the processing device 120 , and the 3D printing device 130 in FIG. 1 . Please refer to the implementation in FIG. 1 . The description of the example is not repeated here.

In this embodiment, the limb guide mechanism 410 includes a body 411 and three positioning holes 412 , 413 and 414 . The positioning holes 412 , 413 and 414 are arranged on the main body 411 . The limb guide mechanism 410 of this embodiment is used to float on the user's hand 420 , which can fix the posture of the user's hand 420 and generate corresponding positioning holes 412 , 413 , 414 of the positioning marks 511, 512, 513, as shown in Figure 5. Then, the point cloud information of the limb surface obtained by the 3D scanning device 110 also includes the positioning marks 511 , 512 and 513 , so that the processing device 120 can perform subsequent operations according to the information of the positioning marks 511 , 512 and 513 . In this way, it is possible to prevent the 3D scanning device 110 from obtaining wrong information due to the shaking of the user's hand.

In this embodiment, the positions of the positioning holes 412 , 413 and 414 are different from each other, and the length of the connecting line between the two positioning holes of the positioning holes 412 , 413 and 414 (for example, the positioning holes 412 and 413 or the positioning holes 413 and 414 ) is less than or equal to the preset distance. That is to say, the connecting line between the positioning holes 412 and 413 and the connecting line between the positioning holes 413 and 414 are respectively smaller than or equal to the predetermined distance. In addition, the above-mentioned preset distance is, for example, 7 centimeters (cm), but the embodiment of the present invention is not limited thereto. Users can adjust the length of the preset distance according to their needs.

In some embodiments, at least one of the positioning holes 412 , 413 and 414 may correspond to a joint point of a limb. It is assumed that the user's hand is used as an example of the limb of the embodiment of the present invention. For example, the positioning hole 412 corresponds to the joint point of the user's thumb adjacent to the tiger's mouth, the positioning hole 413 corresponds to the joint point of the user's wrist, and the positioning hole 414 corresponds to the joint point of the user's hand shaft. In addition, the connection line between the positioning holes 412 and 413 and the connection line between the positioning holes 413 and 414 may form an included angle. Wherein, the angle of the above-mentioned included angle can be adjusted according to requirements, for example, according to the doctor's suggestion.

In addition, the user can also further confirm whether the length of the connection between the two positioning holes of the positioning holes 412, 413 and 414 (for example, the positioning holes 412 and 413 or the positioning holes 413 and 414) is less than or equal to the preset distance to confirm whether to adjust Locate the location of the hole. For example, when the length of the connecting line between the positioning holes 412 and 413 is less than or equal to the predetermined distance, the positions of the positioning holes 412 and 413 remain corresponding to the joint points of the limbs, that is, the positioning hole 412 still corresponds to the user's thumb The locating hole 413 is still corresponding to the joint point of the user's wrist, which is adjacent to the joint point of the palm's mouth. When the length of the connection between the positioning holes 413 and 414 is not less than or equal to the predetermined distance, that is, the length of the connection between the positioning holes 413 and 414 is greater than the predetermined distance, one of the positioning holes 413 and 414 is adjusted according to the predetermined distance. s position. For example, the position of the positioning hole 414 is adjusted, that is, the position of the positioning hole 414 is adjusted to the position between the user's wrist and the hand shaft, and the adjusted position of the positioning hole 414 can be as shown in FIG. 5 . The position of the positioning hole 414 of the limb guide mechanism 410.

In addition, the number of the positioning holes of the limb guide mechanism 410 is three as an example, but is not intended to limit the embodiment of the present invention. The number of the positioning holes of the limb guide mechanism 410 can be four or more, and the positions of the four or more positioning holes are also different from each other, and the same technical effect can still be achieved. The setting method of the four or more positioning holes can also be set and adjusted as in the above embodiment, so it is not repeated here.

In addition, assuming that the present embodiment uses the user's hand as an example of the limb of the embodiment of the present invention, the electronic system 400 for modeling the assistive device further includes a finger ring (not shown). For example, the finger ring is used to fit on the thumb of the user's hand, so that the point cloud information of the limb surface obtained by the three-dimensional scanning device 110 also includes the information corresponding to the finger ring. Then, the processing device 120 can use the known information corresponding to the ring as a basis for modifying and adjusting the selected preset digital model of the assistive device. In this way, the accuracy and quality of the generation of the wearable auxiliary device can be further increased.

FIG. 6 is a flowchart of a method for modeling an assistive device according to another embodiment of the present invention. Please refer to FIGS. 3 to 6 at the same time. In this embodiment, the user's hand is used as an example of the limb of the embodiment of the present invention. In step S602 , at least three positioning marks 511 , 512 , 513 corresponding to the at least three positioning holes 412 , 413 , 414 on the limb guide mechanism 410 are generated on the limb through the limb guide mechanism 410 . In step S604 , the processing device 120 obtains the limb surface point cloud information of the corresponding limb, as shown in FIG. 3 . The above body surface point cloud information is generated by scanning the user's body through the 3D scanning device 110 , and the body surface point cloud information includes three positioning marks 511 , 512 and 513 .

In step S606, the processing device 120 turns the point cloud information of the limb surface to a preset position according to the at least three positioning marks 511, 512, 513, for example, the position corresponding to the mark 310 shown in FIG. 3 . That is to say, when the position of the point cloud information on the limb surface generated by the three-dimensional scanning device 110 is different from the position in FIG. 3 , the processing device 120 can use the coordinate system defined by the positioning marks 511 , 512 and 513 to locate the limb surface The point cloud information is quickly turned to the position marked 310 in FIG. 3 , such as the position where the palm surface of the user's hand is exposed. In this way, the localization of the point cloud information on the body surface can be accelerated. The above-mentioned preset position is an embodiment of the present invention, and is not used to limit the embodiment of the present invention. The user can also adjust the form of the preset position according to his needs, such as the exposed position of the palm back of the user's hand, etc., all of which can achieve the same effect.

In step S608, the processing device 120 generates a plurality of learning data according to the point cloud information of the body surface. Further, the processing device 120, for example, rotates the point cloud information of the body surface to a plurality of different angles and performs proportional scaling, so as to sequentially learn a plurality of data. In addition, the method of generating a plurality of learning materials in step S608 is the same as that in step S204 in FIG. 2 , and the description of step S204 in FIG. 2 can be referred to, so it will not be repeated here.

In step S610, the processing device 120 processes the learning data through a neural-like network to generate a digital model of the assistive device. That is to say, the processing device 120 processes the above learning data through, for example, a convolutional neural network to generate a corresponding digital model of a limb, and draws a digital model of the assistive device corresponding to the digital model of the limb according to the digital model of the limb , to generate a digital model of the assistive device.

In step S612, an auxiliary device is generated according to the digital model of the auxiliary device through the three-dimensional printing device 130. In this way, the wearable assistive device can be created quickly, the design quality of the assistive device can be improved, and the production cycle of the customized assistive device can be shortened.

FIG. 7 is a flowchart of a method for modeling an assistive device according to another embodiment of the present invention. Please refer to FIGS. 3 to 5 and 7 at the same time. In this embodiment, the user's hand is used as an example of the limb of the embodiment of the present invention. In step S702 , a limb guide mechanism 410 is provided, wherein the limb guide mechanism 410 is provided with at least three positioning holes 412 , 413 , and 414 , as shown in FIGS. 4 and 5 . In step S704 , at least three positioning marks 511 , 512 , 513 corresponding to the at least three positioning holes 412 , 413 , 414 are generated on the limb through the limb guide mechanism 410 , as shown in FIG. 5 .

In step S706 , the limb is scanned to obtain point cloud information of the limb surface, and the point cloud information of the limb surface includes at least three positioning marks 511 , 512 and 513 . The above body surface point cloud information is generated by scanning the user's body through the 3D scanning device 110 , and the body surface point cloud information includes three positioning marks 511 , 512 and 513 .

In step S708, the processing device 120 obtains a plurality of preset digital models of assistive devices from the database. That is to say, the processing device 120 can obtain all the digital models of the preset assistive devices in the database, so as to compare the digital models of the preset assistive devices with the point cloud information of the limb surface, and then confirm whether there is a point cloud corresponding to the limb surface A digital model of the default assistive device for information.

In step S710, the processing device 120 superimposes the point cloud information of the limb surface and the digital model of the preset assistive device according to the at least three positioning marks 511, 512, 513 of the point cloud information of the limb surface. For example, the processing device 120 can process the point cloud information of the limb surface and the digital model of the preset assistive device according to the coordinate system defined by the positioning marks 511 , 512 and 513 and the preset coordinate system defined by the digital model of the preset assistive device. Position and align as shown in Figure 8.

In FIG. 8 , the point cloud information 810 of the limb surface and the digital model 820 of the preset assistive device have been adjusted to the positions corresponding to the same coordinate system. Next, the processing device 120 combines the point cloud information 810 of the limb surface with the digital model 820 of the preset assistive device to form a superimposed model 830 as shown in FIG. 8 . In this way, the superimposition speed of the point cloud information of the limb surface and the digital model of the preset aids can be accelerated.

In step S712, a preset digital model of an assistive device that matches the point cloud information of the limb surface is selected. That is to say, after the processing device 120 superimposes the point cloud information of the limb surface with the preset digital model of the assistive device, the processing device 120 can obtain a plurality of superimposed models as described above, and select from the superimposed models. The digital model of the preset assistive device matched with the point cloud information of the limb surface.

In step S714 , the processing device 120 generates a digital model of the assistive device according to the point cloud information of the limb surface and the matched preset digital model of the assistive device. That is to say, the processing device 120 can adjust the appearance and shape of the matching digital model of the preset assistive device according to the point cloud information of the body surface, so that the appearance shape of the digital model of the preset assistive device can conform to the size of the point cloud information on the surface of the body. The adjusted preset digital model of the assistive device is used as the digital model of the assistive device.

In step S716, through the three-dimensional printing device 130, the auxiliary device is generated according to the digital model of the auxiliary device. In this way, a digital model of the wearable assistive device can be quickly established, the design quality of the assistive device can be improved, and the production cycle of the customized assistive device can be shortened.

FIG. 9 is a detailed flowchart of step S714 in FIG. 7 . Please also refer to Figures 3~5 and 7~9. In step S902 , the processing device 120 confirms the correspondence between the first shape corresponding to the point cloud information of the limb surface and the second shape corresponding to the matching preset assistive device digital model. That is to say, the processing device 120 confirms the correspondence between the point cloud information of the limb surface and the appearance shape of the matching digital model of the preset assistive device, as a basis for adjusting the digital model of the preset assistive device.

. In step S904, when it is confirmed that the corresponding relationship is that the second shape is greater than the first shape, the processing device 120 adjusts the second shape of the preset digital model of the assistive device to match the first shape through the first algorithm, so as to generate the digital assistive device Model. In this embodiment, the first algorithm is, for example, Boolean operation subtraction. That is to say, when the size of the appearance shape of the matching preset digital model of the assistive device is larger than the size of the appearance shape of the point cloud information of the limb surface, the processing device 120 calculates the difference between the sizes of the appearance shapes, for example, through Boolean operation subtraction. , and according to the calculated difference, the second shape of the preset digital model of the assistive device is adjusted to conform to the first shape, so as to generate the digital model of the assistive device.

In step S906, when it is confirmed that the corresponding relationship is that one part of the second shape is larger than the first shape and another part of the second shape is smaller than the first shape, or the corresponding relationship is that the second shape is smaller than the first shape, the processing device 120 transmits the The second algorithm adjusts the second shape of the preset digital model of the assistive device to conform to the first shape, so as to generate the digital model of the assistive device. In this embodiment, the second algorithm is, for example, a multi-layer contour method.

That is to say, when the size of the appearance shape of the matching digital model of the preset auxiliary device is partially or totally smaller than the size of the appearance shape of the point cloud information of the limb surface, the processing device 120, for example, through the multi-layer outline method, The appearance shape with the digital model is divided into multi-layer sections to obtain the contour lines corresponding to the multi-layer sections, as shown in Figure 10. Next, the processing device 120 adjusts the contours of the multi-layered cross-sections according to the point cloud information of the limb surface, so that the second shape of the preset digital model of the assistive device is adjusted to conform to the first shape, so as to generate the digital model of the assistive device. In this way, a digital model of the wearable assistive device can be quickly established, the design quality of the assistive device can be improved, and the production cycle of the customized assistive device can be shortened.

FIG. 11 is a flowchart of a method for modeling an assistive device according to another embodiment of the present invention. Please also refer to Figures 3~5 and 7~11. In this embodiment, steps S702 - 716 are the same as or similar to steps S702 - 716 in FIG. 7 . Reference can be made to the description of the embodiment in FIG. 7 , and details are not repeated here.

Following step S710, in step S1102, the processing device 120 confirms whether the point cloud information of the limb surface matches at least one of the preset digital models of assistive devices. That is, the processing device 120 will confirm whether there is a suitable digital model of the preset assistive device in the database.

When it is confirmed that the point cloud information of the limb surface matches at least one of the digital model of the preset assistive device, it means that the difference between the point cloud information of the limb surface and the digital model of the preset assistive device is small, and there is a suitable digital model of the preset assistive device in the database. can use. Next, go to step S712, select a preset digital model of an assistive device that matches the point cloud information of the limb surface, and execute steps S714-S716 to generate a suitable assistive device.

When it is confirmed that the point cloud information of the limb surface does not match the digital model of the preset assistive device, it means that the point cloud information of the limb surface is quite different from the digital model of the preset assistive device, and there is no suitable digital model of the preset assistive device in the database. . Next, in step S1104, the processing device 120 may process the point cloud information on the body surface through a neural-like network to generate a digital model of the assistive device. That is to say, the processing device 120 can generate a suitable digital model of the assistive device through steps S604-S610 in FIG. 6, that is, step S1104 in this embodiment may include steps S604-S610 in FIG.

In the aforementioned embodiment, when the above step S1102 confirms that the point cloud information of the limb surface matches at least one of the preset digital model of the assistive device, the shape of the predetermined digital model of the assistive device can be adjusted by the Boolean operation subtraction method or the multi-layer contour method. , to generate a digital model of the assistive device. This embodiment is only an example of the embodiment of the present invention, but the embodiment of the present invention is not limited thereto. In future assistive device modeling applications, the processing device 120 can use the above-mentioned neural network-like learning method to replace the Boolean operation subtraction method or the multi-layer contour method to generate the assistive device digital model, and can also achieve similar assistive device modeling. model effect.

To sum up, the assistive device modeling method and the limb guide mechanism disclosed in the embodiments of the present invention generate a plurality of learning data according to the point cloud information of the limb surface, and process the learning data through a neural network. to generate digital models of assistive devices. In addition, the embodiments of the present invention can also provide a limb guide mechanism with at least three positioning holes, and generate at least three positioning marks corresponding to the at least three positioning holes on the limb through the limb guide mechanism. Next, scan the limb to obtain the point cloud information of the limb surface including at least three positioning marks, and according to the at least three positioning marks of the point cloud information of the limb surface, compare the point cloud information of the limb surface with a plurality of preset points obtained in the database. The assistive device digital model is set for superimposition, the preset assistive device digital model that matches the point cloud information of the limb surface is selected, and then the assistive device digital model is generated according to the limb surface point cloud information and the matching preset assistive device digital model. In this way, an auxiliary device digital model of the wearable auxiliary device can be established effectively and quickly, the design quality of the auxiliary device can be improved, and the production cycle of the customized auxiliary device can be shortened.

Although the present invention is disclosed above by the embodiments, it is not intended to limit the scope of the present invention. Anyone with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be determined by the scope of the appended patent application.

100, 400: Electronic systems for assistive device modeling 110: 3D scanning device 120: Processing device 130: 3D Printing Device 310, 810: Limb surface point cloud information 320: Rotation direction 410: Limb Guide Mechanism 411: Ontology 412, 413, 414: positioning holes 420: Hands 511, 512, 513: positioning marks 820: Preset digital model of aids 830: Superimposed Models S202~S208, S602~S612, S702~S716, S902~S906, S1102~S1104: Steps

In order to have a better understanding of the above-mentioned and other aspects of the present invention, the following specific examples are given and described in detail in conjunction with the accompanying drawings as follows: FIG. 1 is a schematic diagram of an electronic system for modeling an assistive device according to an embodiment of the present invention. FIG. 2 is a flowchart of an assistive device modeling method according to an embodiment of the present invention. FIG. 3 is a schematic diagram of generating a plurality of learning data according to the point cloud information of the body surface according to an embodiment of the present invention. FIG. 4 is a schematic diagram of an electronic system for modeling an assistive device according to an embodiment of the present invention. FIG. 5 is a schematic diagram of the corresponding relationship between the positioning holes of the limb guide mechanism and the positioning marks on the limb according to an embodiment of the present invention. FIG. 6 is a flowchart of a method for modeling an assistive device according to another embodiment of the present invention. FIG. 7 is a flowchart of a method for modeling an assistive device according to another embodiment of the present invention. FIG. 8 is a schematic diagram of the correspondence between the point cloud information of the limb surface, the digital model of the preset assistive device and the superimposed model according to the present invention. FIG. 9 is a detailed flowchart of step S714 in FIG. 7 . FIG. 10 is a schematic diagram of dividing the appearance shape of the digital model of the matching preset aid into multi-layer sections according to the present invention. FIG. 11 is a flowchart of a method for modeling an assistive device according to another embodiment of the present invention.

S202~S208: Steps

Claims (13)

An assistive device modeling method, comprising: obtaining point cloud information of a limb surface corresponding to a limb; generating a plurality of learning data according to the point cloud information of the limb surface; A limb model is generated, and according to the limb model, a digital model of an assistive device corresponding to the limb model is drawn to generate the digital model of the assistive device. The assistive device modeling method as described in item 1 of the scope of the application, wherein the step of generating the learning data according to the point cloud information of the limb surface includes: rotating the point cloud information of the limb surface to a plurality of different angles and performing equal scaling scaling to generate these learning materials. The assistive device modeling method as described in item 1 of the scope of the application, further comprising: generating at least three positioning marks on a limb corresponding to at least three positioning holes on the limb guide mechanism through a limb guide mechanism . The assistive device modeling method as described in item 3 of the scope of the patent application, wherein the step of generating the learning data according to the point cloud information of the limb surface includes: turning the point cloud information of the limb surface according to the at least three positioning marks to a preset position; and rotating the body surface point cloud information to a plurality of different angles and performing proportional scaling to generate the learning data. The assisting device modeling method as described in item 1 of the scope of the patent application further comprises: generating an assisting device according to the digital model of the assisting device through a three-dimensional printing device. The assisting device modeling method as described in item 1 of the scope of the application, wherein the step of obtaining point cloud information corresponding to the surface of the limb includes: scanning the limb through a three-dimensional scanning device to obtain the surface of the limb corresponding to the limb Point cloud information. A method for modeling an auxiliary device, comprising: providing a limb guide mechanism, wherein the limb guide mechanism is provided with at least three positioning holes; through the limb guide mechanism, a limb corresponding to the at least three positioning holes is generated on a limb at least three positioning marks; scanning the limb to obtain point cloud information of a limb surface, and the point cloud information of the limb surface includes the at least three positioning marks; obtaining a plurality of preset digital models of assistive devices from a database; For the at least three positioning marks of the point cloud information of the limb surface, superimpose the point cloud information of the limb surface with the digital models of the preset assistive devices; select a preset assistive device digital model that matches the point cloud information of the limb surface a model; and generate a digital model of an assistive device according to a first shape corresponding to the point cloud information of the limb surface and a second shape corresponding to the matching digital model of the preset assistive device; wherein, according to the point cloud of the limb surface The first shape corresponding to the information and the second shape corresponding to the matching preset digital model of the assistive device, and the step of generating the digital model of the assistive device includes: confirming that the first shape corresponding to the point cloud information of the limb surface and the corresponding digital model a corresponding relationship of the second shape corresponding to the matching digital model of the preset aid; When it is confirmed that the corresponding relationship is that the second shape is larger than the first shape, a first algorithm is used to adjust the second shape of the preset digital model of the assistive device to match the first shape, so as to generate the digital model of the assistive device ; and when confirming that the corresponding relationship is that one part of the second shape is larger than the first shape and another part of the second shape is smaller than the first shape, or the corresponding relationship is that the second shape is smaller than the first shape, through A second algorithm for adjusting the second shape of the preset digital model of the assistive device to conform to the first shape to generate the digital model of the assistive device. The assistive device modeling method as described in item 7 of the scope of the patent application further comprises: confirming whether the point cloud information on the surface of the limb matches at least one of the preset digital models of the assistive device; when confirming that the point cloud information on the limb surface matches the When at least one of the preset assistive digital models matches, select the predetermined assistive digital model matching the point cloud information of the limb surface; and when confirming that the point cloud information of the limb surface does not match the predetermined assistive digital models When the model is matched, the point cloud information on the surface of the limb is processed through a neural network to generate a limb model, and according to the limb model, a digital model of the assistive device corresponding to the limb model is drawn to generate the digital model of the assistive device Model. The assistive device modeling method as described in item 8 of the scope of application, wherein the point cloud information on the surface of the limb is processed through the neural network to generate the limb model, and according to the limb model, the corresponding limb is drawn The assisting device digital model of the model, the step of generating the assisting device digital model includes: generating a plurality of learning data according to the point cloud information of the limb surface; and The learning data are processed through the neural network to generate the limb model, and according to the limb model, the assisting device digital model corresponding to the limb model is drawn to generate the assisting device digital model. The assistive device modeling method as described in item 9 of the scope of the patent application, wherein the step of generating the learning data according to the point cloud information of the limb surface includes: turning the point cloud information of the limb surface according to the at least three positioning marks to a preset position; and rotating the body surface point cloud information to a plurality of different angles and performing proportional scaling to generate the learning data. The assistive device modeling method as described in claim 7, wherein the first algorithm is a Boolean operation subtraction method, and the second algorithm is a multi-layer contour method. The assisting device modeling method as described in item 7 of the scope of the patent application further comprises: generating an assisting device according to the digital model of the assisting device through a three-dimensional printing device. The assistive device modeling method as described in item 7 of the scope of the patent application, wherein the step of scanning the limb to obtain point cloud information on the surface of the limb comprises: scanning the limb through a three-dimensional scanning device to obtain the limb Surface point cloud information.

Images