Search

KR-20260063828-A - APPARATUS FOR SORTING FRUIT USING PICKUP ROBOT

KR20260063828AKR 20260063828 AKR20260063828 AKR 20260063828AKR-20260063828-A

Abstract

The present invention relates to a fruit sorting device using a pickup robot, comprising: a conveyor belt through which a target fruit moves along a first direction; a first camera module for photographing the target fruit moving along the conveyor belt in the first direction; a second camera module for photographing the target fruit moving along the conveyor belt in the first direction at a rear end position of the first camera module in the first direction; a rotating pickup robot for picking up the target fruit between the first camera module and the second camera module, rotating it, and re-seating it on the conveyor belt so as to change the view direction of the target fruit photographed by the first camera module and the second camera module; a first sorting pickup robot for picking up the target fruit that has passed through the second camera module; and a sorting control terminal for analyzing the fruit images captured by the first camera module and the second camera module to determine whether the target fruit is defective, and controlling the first sorting pickup robot so that the target fruit is classified based on the determination result. Through this, a fruit sorting device using a pickup robot is provided, which classifies target fruits based on images of the target fruits and selects the target fruits using a pickup robot, such as a collaborative robot, during the classification process, thereby minimizing damage to the target fruits.

Inventors

  • 권경주

Assignees

  • 주식회사 지에프

Dates

Publication Date
20260507
Application Date
20241031

Claims (7)

  1. In a fruit sorting device using a pickup robot, A conveyor belt through which the target fruit moves along the first direction; A first camera module for photographing a target fruit moving in the first direction along the conveyor belt; A second camera module that photographs a target fruit moving in the first direction along the conveyor belt at a rear end position in the first direction of the first camera module; A rotating pickup robot that picks up a target fruit between the first camera module and the second camera module and rotates it to re-settle it on the conveyor belt, so that the view direction of the target fruit being photographed by the first camera module and the second camera module is changed; A first sorting pickup robot that picks up target fruit that has passed through the second camera module; A fruit sorting device using a pickup robot, characterized by including a sorting control terminal that analyzes fruit images captured by the first camera module and the second camera module to determine whether a target fruit is defective, and controls the first sorting pickup robot to classify the target fruit based on the determination result.
  2. In paragraph 1, It further includes a second sorting pickup robot that picks up target fruit between the first camera module and the rotary pickup robot; A fruit sorting device using a pickup robot, characterized in that the sorting control terminal controls the second sorting pickup robot to pick up and sort the target fruit when the target fruit is determined to be defective based on a fruit image captured by the first camera module.
  3. In paragraph 1, The above-mentioned first sorting pickup robots are arranged in multiple numbers; The above screening control terminal One of the plurality of first sorting pickup robots is controlled to pick up and sort a target fruit identified as defective, and A fruit sorting device using a pickup robot, characterized by controlling the remainder of the plurality of first sorting pickup robots to pick up and sort target fruits determined to be good quality.
  4. In paragraph 3, The above screening control terminal Classify target fruits identified as good quality by quality grade, and A fruit sorting device using a pickup robot, characterized by controlling the remainder of the plurality of first sorting pickup robots to pick up and sort each according to good grade.
  5. In paragraph 3, The above screening control terminal An artificial intelligence-based sorting AI model that receives fruit images captured by the first camera module and the second camera module, respectively, and infers whether the target fruit is defective; A fruit sorting device using a pickup robot, characterized by including a sorting control unit that controls the sorting pickup robot so that target fruits moving along the conveyor belt are sorted according to the inference result of the sorting AI model.
  6. In paragraph 5, The sorting control terminal further includes a target tracker that tracks target fruits inferred as defective by the sorting AI model during the movement process of the conveyor belt; A fruit sorting device using a pickup robot, characterized in that the sorting control unit controls the sorting pickup robot based on the tracking result of the target tracker.
  7. In paragraph 5, The above-mentioned screening control terminal further includes a terminal communication unit that communicates with the outside; A fruit sorting device using a pickup robot, characterized in that the sorting control unit connects to an external fruit sorting support server through the terminal communication unit to download or update the sorting AI model.

Description

Fruit sorting apparatus using a pickup robot The present invention relates to a fruit sorting device using a pickup robot, and more specifically, to a fruit sorting device using a pickup robot capable of automatically classifying defective fruits and good fruits based on an image of a target fruit to be sorted. The sorting of fruits, such as apples, is a process that commercializes harvested fruits. By sorting harvested fruits fresh and uniformly, it enhances the commercial value of agricultural products and possesses the characteristic of increasing added value. In particular, fruit sorting is required because there are various factors that determine the quality of each individual fruit, such as size, weight, shape, color, and the presence or absence of blemishes. Since sorting criteria differ not only by fruit type but also by variety, classification by grade is necessary accordingly. Therefore, uniform sorting and packaging must be carried out to enhance marketability. Generally, the fruit sorting process involves workers classifying fruits moving along a conveyor belt by grade, but recently, fruit sorting devices that automatically classify and sort fruits by size or weight are widely used. While there are methods to classify target fruits by grade based on size or weight, there are also methods to sort them into grades based on specific standards, even for fruits of the same size or weight. Generally, the detection of defects in fruit is performed based on images captured by a camera. For example, a fruit sorting device disclosed in Korean Patent Publication No. 10-2020-0007385 uses a set of three cameras to photograph a target fruit and detects defects based on images captured in three directions. The fruit sorting device disclosed in the aforementioned Korean published patent detects defects based on images taken from three directions, thereby resolving the problem of not being able to detect defects on the opposite surface from images taken from only one direction, but there is a problem in that the number of cameras installed increases in that three cameras are used as a set for shooting. Meanwhile, when the defect detection process is completed, the fruits moving along the conveyor belt are classified into defective fruits and good fruits and stored. For example, the fruit sorting device disclosed in Korean Registered Patent Publication No. 10-1686618 is equipped with multiple sorting conveyors branching off from the main conveyor, and the fruits are classified by branching to specific sorting conveyors according to sorting criteria and dropping into shipping boxes, etc. However, while the aforementioned classification process may result in minimal damage to relatively hard fruits like apples during transport or dropping, fruits such as peaches and pears are susceptible to damage from minor external impacts. In particular, as time passes after even minor damage, the damaged area becomes visually apparent, raising concerns that fruits classified as good quality may be delivered to consumers in a damaged state. FIG. 1 is a diagram showing the configuration of a fruit sorting support system according to one embodiment of the present invention, and FIG. 2 is a diagram showing an example of the configuration of a fruit sorting support server according to an embodiment of the present invention, and FIGS. 3 and 4 are drawings for explaining the types of classification models, learning processes, and classification processes according to an embodiment of the present invention, and FIG. 5 is a diagram showing an example of the configuration of a fruit sorting device using a pickup robot according to a first embodiment of the present invention, and FIG. 6 is a control block diagram of a fruit sorting device using a pickup robot according to a first embodiment of the present invention, and FIG. 7 is a diagram showing an example of the configuration of a fruit sorting device using a pickup robot according to a second embodiment of the present invention. The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention. The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions,