KR-102963932-B1 - DISH DATABASE COLLECTION APARATUS FOR DISH LEARNING AND RECOGNITION

KR102963932B1KR 102963932 B1KR102963932 B1KR 102963932B1KR-102963932-B1

Abstract

A database collection device for learning and recognizing tableware is disclosed. The database collection device for learning and recognizing tableware according to the present invention is characterized by comprising: an arc-shaped rail; a housing slidably coupled to the arc-shaped rail; a camera coupled to the housing such that the shooting direction is parallel to the radius of curvature of the arc-shaped rail; and a turntable provided on the side of the camera's shooting direction, on which tableware is placed. According to the present invention, a database collection device for learning and recognizing tableware can be provided, wherein the camera generates image data of various types of tableware at an accurate position and angle to build a reliable database for various types of tableware.

Inventors

윤성조
박정우
엄태영
황정환
김효곤
최영호
정구봉

Assignees

한국로봇융합연구원

Dates

Publication Date: 20260511
Application Date: 20231108

Claims (9)

A turntable on which tableware is placed on the surface; A camera that photographs tableware placed on the upper surface of the above-mentioned turntable; Housing to which the above camera is attached; and It includes an arc-shaped rail spaced apart from the above-mentioned turntable at a certain distance, and The above arc-shaped rail is, A first arc-shaped rail surface with a normal facing toward the turntable; and It includes a second arc-shaped rail surface having the same center of curvature as the first arc-shaped rail surface and a normal facing the opposite side of the turntable, The above housing slides on the arc-shaped rail while in contact with the first arc-shaped rail surface and the second arc-shaped rail surface, and As the camera is coupled to the housing such that its shooting direction is parallel to the radius of curvature of the arc-shaped rail, the turntable is positioned on the side of the camera's shooting direction during the process in which the housing slides along the arc-shaped rail while in contact with the first arc-shaped rail surface and the second arc-shaped rail surface. Database collection device for tableware learning and recognition.
A turntable on which tableware is placed on the surface; A camera that photographs tableware placed on the upper surface of the above-mentioned turntable; An arc-shaped rail spaced apart from the above turntable at a certain distance; and It includes a housing to which the camera is coupled so as to be slidably movable on the above-mentioned arc-shaped rail, and The above arc-shaped rail includes a first arc-shaped rail surface having the same center of curvature but with a normal pointing toward the turntable, and a second arc-shaped rail surface with a normal pointing toward the opposite side of the turntable. The above housing is, A first surface in contact with the first arc-shaped rail surface; and It includes a rolling wheel that contacts the second arc-shaped rail surface on the opposite side from the first surface based on the arc-shaped rail, As the camera is coupled to the housing such that its shooting direction is parallel to the radius of curvature of the arc-shaped rail, the turntable is positioned on the side of the camera's shooting direction during the process in which the housing slides along the arc-shaped rail while in contact with the first arc-shaped rail surface and the second arc-shaped rail surface. Database collection device for tableware learning and recognition.
In paragraph 2, The above housing is, A lower housing that forms the first surface and to which the camera is fixed; and The above-mentioned rolling wheel is rotatably coupled, and the upper housing is coupled to the lower housing in a vertical direction of the first surface, comprising Database collection device for tableware learning and recognition.
In paragraph 2, The first arc-shaped rail surface and the first surface maintain a state of contact due to the elastic force of the rolling wheel. Database collection device for tableware learning and recognition.
Arc-shaped rail; A housing slidably coupled to the above arc-shaped rail; A camera coupled to the housing such that the shooting direction is parallel to the radius of curvature of the arc-shaped rail; and It includes a turntable provided on the shooting direction side of the above camera, on which tableware is placed on the upper surface, and The above arc-shaped rail is, A first arc-shaped rail surface with a normal facing toward the turntable; and It includes a second arc-shaped rail surface having the same center of curvature as the first arc-shaped rail surface and a normal facing the opposite side of the turntable, As the camera is coupled to the housing such that its shooting direction is parallel to the radius of curvature of the arc-shaped rail, the turntable is positioned on the side of the camera's shooting direction during the process in which the housing slides along the arc-shaped rail while in contact with the first arc-shaped rail surface and the second arc-shaped rail surface. Database collection device for tableware learning and recognition.
In paragraph 5, The above arc-shaped rail includes a first arc-shaped rail and a second arc-shaped rail spaced apart from each other in a direction perpendicular to the radius of curvature, and The camera is coupled to the housing between the first arc-shaped rail and the second arc-shaped rail, Database collection device for tableware learning and recognition.
In paragraph 5, A device comprising an actuator that slides the housing on the above arc-shaped rail, Database collection device for tableware learning and recognition.
In Paragraph 7, The above actuator is, A pair of pulleys provided at both ends of the above arc-shaped rail; The above housing is coupled, and a wire is hung on the above pair of pulleys; A motor that rotates one of the above pair of pulleys; and A controller comprising a controller that controls the above motor, Database collection device for tableware learning and recognition.
In Paragraph 7, The above actuator is, The above housing is joined, and a wire having a plurality of beads joined; A first pulley provided at one end of the above arc-shaped rail and on which a wire is caught; A second pulley provided at the other end of the above arc-shaped rail, wherein the bead is inserted between the teeth; A motor that rotates the second pulley; and A controller comprising a controller that controls the above motor, Database collection device for tableware learning and recognition.

Description

DISH DATABASE COLLECTION APARATUS FOR DISH LEARNING AND RECOGNITION The present invention relates to a database collection device for learning and recognizing tableware, and more specifically, to a database collection device for learning and recognizing tableware configured to create a database of information about tableware. The development of IT technology driven by the Fourth Industrial Revolution has brought a new trend called Foodtech to the food and beverage industry. Foodtech, a term combining "Food" and "Technology," refers to a new industry that integrates the food industry with IT technologies such as big data. Kiosks are already taking orders in place of clerks at many food and beverage establishments, and recently, robots responsible for serving food have also been introduced. Serving robots are indoor autonomous robots that replace store employees in delivering food; they navigate by recognizing their location and obstacles using mapping technology that learns the space to create maps, as well as sensors. However, for now, it only takes over the simple task of moving food, and tasks such as clearing away customers' empty dishes still remain the responsibility of employees. For a robot to collect dishes left behind by customers, it must first accurately recognize the tableware on the table through a camera; to achieve this, the process of creating a database of the tableware information must be performed beforehand. In this regard, Korean Registered Patent Publication No. 2210533 (hereinafter referred to as the 'prior art') discloses a 3D scanning device for an object. The 3D scanning device of the prior art may include an object table having an object mounting portion capable of mounting an object and a table rotation portion formed inside the object mounting portion and rotating in one direction; an image collection device having a camera portion including one or more cameras installed at regular intervals in a frame, capturing an object through the camera portion to collect images and information of the object, and a frame adjustment portion installed between the frames to adjust the length of the frame; and a server that adjusts the angle of the camera portion and the length of the frame to adjust according to the size of the object and 3D models one or more images of the object provided by the image collection device. The 3D scanning device of the prior art has the advantage of being able to perform 3D modeling by collecting images of an object with minimal number of cameras for collecting images of an object and by calculating and adjusting the optimal angle and position of the camera for capturing the object. In order for a robot to accurately recognize tableware placed on a table through a camera, the camera must photograph various types of tableware under identical position and angle conditions when creating a database of tableware information. However, in order for multiple cameras to photograph objects from various angles, the angles of the cameras and the lengths of the frames must be adjusted through a shaft adjustment unit, a motor, a frame adjustment unit, and a length adjustment unit. Therefore, if an operating error occurs in one or more of the components of the shaft adjustment unit, the motor, the frame adjustment unit, and the length adjustment unit during this process, the reliability of the database information decreases, and consequently, the recognition rate of tableware is lowered. Various types of tableware are diverse and inconsistent in terms of type, shape, and color, and there is a problem in standardizing these characteristics due to leftover food residue. Therefore, it is necessary to establish a reliable database of tableware by having a camera generate image data of the tableware from precise positions and angles. FIG. 1 is a perspective view showing a database collection device for tableware learning and recognition according to an embodiment of the present invention. FIG. 2 is a side view of a database collection device for tableware learning and recognition according to an embodiment of the present invention. Figure 3 is a drawing showing part A of Figure 2. FIG. 4 is a front view of a camera module according to an embodiment of the present invention. Figure 5 is a perspective view of part B of Figure 2, showing an actuator. Figure 6 is a perspective view of part C of Figure 2. Figure 7 is a cross-sectional view of part A of Figure 2. FIG. 8 is a perspective view of part B of FIG. 2, showing the second pulley and wire. Figure 9 is a drawing showing the FF' cross-section of Figure 8. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. However, in describing the present invention, descriptions of already known functions or configurations will be omitted in order to clarify the gist of the present invention. FIG. 1 is a perspective view showing a database collection device (10) for lear