KR20110137532A - Automatic recongnition and performance processing system for products - Google Patents

Abstract

PURPOSE: An automatic product recognition processing method is provided to automatically store product information and to count the stock of a product by automatically recognizing the production situation of a specific product. CONSTITUTION: Product information is inputted(S100). The inputted production information is transmitted to a POP terminal(S120). The manufactured product is recognized through shape comparison(S130). The recognized data is converted into data for heterogeneous network(S140). The converted information is transmitted to a gateway or a server(S150).

Description

AUTOMATIC RECONGNITION AND PERFORMANCE PROCESSING SYSTEM FOR PRODUCTS}

The present invention relates to automatic product identification and performance processing system, and automatically recognizes the production status of a specific product, and automatically stores all the information related to the product as well as whether the product is automatically counted and the production of the product inventory count To a processing system.

In general, in the case of producing a product, the manager inputs information on the product, such as necessary materials and wages.

For example, when a production-related department such as production management sets a production quantity of a corresponding product and gives a production instruction, it grasps a required material and counts the input of the corresponding material.

Since such a series of production information must be input by the manager one by one, a mistake of an unskilled worker or an operator working in a production line frequently causes incorrect data to be input.

In addition, since all the production information must be input by the manager or the producer, there is a problem that the work efficiency is reduced.

Accordingly, an object of the present invention is to solve the problems of the prior art as described above, by automatically taking information of the product by taking an image of the product, whether the product is automatic to count the inventory of the product and It is to provide a performance processing system.

According to a feature of the present invention for achieving the above object, the product information input step of inputting information on the product; A production step of producing a product; A POP transmission step of transmitting the production related information input in the product information input step to a POP terminal; A product recognition step of recognizing the product by the shape comparison of the product actually produced in the production step; A data conversion step of converting the result data of the POP transmission step and the product recognition step into data for heterogeneous networks; A gateway transmission step or server transmission step of transmitting the information passed through the data conversion step to a gateway or a server; A judging step of judging whether the information in both steps is identical by comparing the production information by the POP transmission step with the product information recognized in the product recognition step; A production product count step of increasing the inventory count of the corresponding product if the information in the POP transmission step and the product recognition step match each other as a result of the determination in the determination step; A back calculation step of inversely calculating part information on a product produced by back ordering when the information in the POP transmission step and the product recognition step coincide with each other as a result of the determination in the determination step; A set product count step of increasing the production stock count of the POP set product in addition to the back calculation step; And a storage step of storing information on the produced product.

In the automatic product recognition and performance processing system according to the present invention, information on the current product produced by the image information of the previously learned product is automatically determined and stored. Therefore, there is an advantage in that wrong information input by an unskilled person is reduced as in the prior art.

In addition, the present invention is configured to automatically check the learned information and the information in the image to automatically count the inventory data and the product count. Therefore, since a separate input operation is unnecessary, there is an effect that the work efficiency is improved.

1 is a flow chart showing the configuration of a preferred embodiment of the automatic product recognition and performance processing system according to the present invention.
Figure 2 is a flowchart showing the flow of the learning step used in the embodiment of the present invention.
Figure 3 is a flow chart showing the specific flow of the data conversion step of the embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the automatic product recognition and performance processing system according to the present invention will be described in detail.

1 is a flow chart showing an embodiment of the automatic product recognition and performance processing system according to the present invention.

As shown in the figure, the automatic product recognition and performance processing system according to the present invention includes a product information input step (S100) for inputting information on a product, a production step (S110) for producing a product, and the product information. POP recognition step (S120) for transmitting the production-related information input in the input step (S100) to the POP terminal, and product recognition step for recognizing the product itself by comparing the shape of the actual product produced in the production step (S110) ( S130), a data conversion step (S140) for converting data into heterogeneous networks, and a gateway transmission step (S150) or server transmission step (S160) for transmitting the information passed through the data conversion step (S140) to a gateway or a server. And comparing the production information by the POP transmission step (S120) and the product information recognized in the product recognition step (S130) with each other to determine whether a match is determined (S170), and the POP transmission step (S120) and product recognition. If the information in the system (S130) coincide with each other, the product product counting step (S180) for increasing the inventory count of the corresponding product, and back-calculating part information on the product produced by the back ordering (Back Ordering) The reverse calculation step (S190), the set product count step (S200) for increasing the production inventory count of the POP set product, and the storage step (S210) for storing information about the produced product.

In the product information input step (S100), a manager inputs information on a product to be produced. In other words, it is a process of inputting necessary materials and production time required for the production of the product.

The production step (S110) is a process of manufacturing the planned product in the actual line (Line).

The POP transmission step (S120) is a step in which production information is transmitted to the POP terminal when production of the product is in progress. That is, the process of transmitting the production-related information input in the product information input step (S100) to the POP terminal installed next to the facility.

The product recognition step (S130) is a step of recognizing the product itself by comparing the shape of the product actually produced in the production step (S110).

The product recognition step (S130), as shown, preferably proceeds simultaneously with the POP transmission step (S120).

In the product recognition step (S130), the quality is determined by a recognition engine based on Hierarchical Temporal Memory (HTM).

Specifically, as shown, the product recognition step (S130), the shooting process (S132) for taking an image of the actual product produced, and the management process for managing the image taken in the shooting process (S132) ( S134) and a recognition process (S136) for recognizing the product in comparison with the learning information learned in the learning step (S300).

The photographing process (S132) is a process of photographing an image of a product flowing in an actual production line with a camera or the like.

The management process (S134) is a process of storing, deleting, adding, inquiring, etc. of the image photographed in the photographing process (S132).

In the recognition process (S136), in a separate learning step (S300) to use the image data obtained by learning the images of the various products in advance. In addition, the image photographed in the photographing process (S132) is compared with the learning information learned in the learning step (S300), and in this case, the shape of the product is recognized by a recognition engine based on HTM. do. Detailed content of the learning step (S300) will be described in detail below.

The data conversion step (S140) is a step of converting data related to product information such as a product code and a product count in order to transmit data to another model. That is, the process of converting the result data of the POP transmission step (S120) and the product recognition step (S130) into data for networks of different models.

The data conversion step (S140) is performed before the gateway transmission step (S150) and the server transmission step (S160). Specifically, after the data conversion step (S140), the gateway transmission step (S150) and the server transmission step (S160) proceed sequentially.

The gateway transmission step S150 is selectively performed before the server transmission step S160. That is, the gateway transmission step S150 is an intermediate step for the transmission of the long distance data before the server transmission step S160, and is selectively performed as necessary.

As described above, the gateway transmission step S150 is a step of transmitting data to an intermediate transmission device such as a repeater or a concentrator for long distance data transmission. Accordingly, when short or long distance data transmission is possible and when data concentration is unnecessary, such a gateway transmission step S150 may be omitted.

The server transmission step (S160) is a step of transmitting product information to the server, and the data conversion step (S140) may be performed even before the server transmission step (S160). That is, the data conversion step (S140) is preferably executed before the gateway transmission step (S150) and the server transmission step (S160), respectively.

The determining step (S170), by comparing the production information by the POP transmission step (S120) and the product information recognized in the product recognition step (S130) with each other, and determining whether the information in both steps match each other to be.

The production product count step (S180), when the determination result in the determination step (S170), when the information in the POP transmission step (S120) and the product recognition step (S130) match each other, the inventory count of the corresponding product It is a step to increase. In other words, since the production of the product is normally planned correctly, the quantity of the product is increased.

The inverse calculation step (S190), if the information in the POP transmission step (S120) and product recognition step (S130) is the same as the determination result in the determination step (S170), the back ordering (Back Ordering) It is the process of back-calculating the parts information about the produced product.

Such a reverse calculation step (S190) is carried out even if the information in the POP transmission step (S120) and the product recognition step (S130) does not match each other, the POP transmission step (S120) is not performed.

In the reverse calculation step (S190), artificially calculates the material requirements required for the production of the product, the manager directly calculates information on the specific product production, such as material input history and wage cost calculation.

The set product count step (S200), in addition to the back calculation step (S190), is a process of increasing the production inventory count of the POP set product.

In addition, the storage step (S210) is a process of storing information about the product produced by each step as described above. The storage step S210 may be performed between the steps as before the determination step S170.

After such a number of steps are performed, production of one product is complete.

2 is a flowchart showing the flow of the learning step (S300).

As shown in this, the learning step (S300) used in the recognition process (S136), the item setting process (S310) for setting and managing the item to learn, and the shape learning process (S320) for learning the image of the product And, the learning shape management process (S330) for managing the image learned in the shape learning process (S320), the shape test process (S340) for testing the learned image, and the shape test process (S340) Test shape management process (S350) for managing the shape, recognition rate inspection process (S360) for checking the recognition rate level of the image tested by the test shape management process (S350), and determine whether the recognition rate has a predetermined level or more The recognition rate determination step (S370) and so on.

The item setting process (S310) is a process of managing the item to be learned through the addition / deletion / storage / inquiry.

And, although not shown, in the item setting process (S310), the item selection process for selecting the article to be learned, the type determination process for determining whether to continue learning or new learning, the type determination process in the In the case of new learning as a result of the determination, the category formation process of forming a new category, and in the case of the type determination process, in the case of continuing the existing learning, the category selection process of selecting an existing category, and the number of data to be learned for each category It can be divided into detailed processes such as a ratio selection process for selecting a ratio and a folder selection process for selecting a folder for storing image data learned by category.

The shape learning process (S320) is a process of learning by receiving shape data of various products.

The learning shape management process (S330) is a process of managing a learned image, and is a process of classifying and managing by adding / deleting / storing / searching the learned data.

The shape test process S340 is a process of testing the learned image using an actual product. That is, the process of testing whether the product is recognized by receiving an image of a product manufactured in an actual production line and comparing the image with the learned image.

The test shape management process S350 is a process of classifying and managing the test data by adding / deleting / storing / searching the test data by the shape test process S340.

When the test of the actual product is completed as described above, the recognition rate inspection process (S360) is next. The recognition rate inspection process (S360) is a process of checking and digitizing how much the recognition rate level of the image tested by the test shape management process (S350) is.

When the recognition rate level is determined by the test, it is determined whether the recognition rate level is applicable to the actual production line. That is, it is determined whether the recognition rate level derived by the recognition rate inspection process (S360) has a recognition rate of a predetermined level or more, and this step is the recognition rate determination step (S370).

As described above, in the recognition rate determining step (S370), the recognition rate by the test is compared with the preset recognition rate and the recognition rate by the test.

As a result of the determination in the recognition rate determining step S370, when the recognition rate level does not satisfy a predetermined level, the process returns to the shape learning process S320. Thus, learning takes place again.

On the other hand, as a result of the determination in the recognition rate determination step (S370), when the recognition rate level satisfies a predetermined level, the learning step (S300) is terminated, the learning data material is applied to the actual production is used.

3 is a detailed flowchart of the data conversion step (S140).

As shown therein, the data conversion step (S140) includes a method selection process (S410) of selecting a transmission method between heterogeneous types, a synchronization process (S420) of synchronizing communication hardware, and conversion of transmission data according to a protocol. A transmission data conversion process (S430), a transmission process for transmitting data to a receiver (S440), a reception process for receiving data at the receiver (S450), and reception data for converting the received data into received data according to a protocol A conversion process (S460), a packet checking process (S470) for checking whether there is a packet loss in the received data, and a packet retransmission request process (S480) for requesting retransmission of data, if there is a packet loss.

The method selection process (S410) is a process of selecting a transmission method to be used between the transmitter and the receiver. For example, when the data conversion step (S140) is performed before the gateway transmission step (S150), which transmission method is used between POP-MCUs is selected, and the data conversion step (S140) transmits the server. If performed before step S160, it is to select which transmission method to use between the MCU and the server.

The synchronization process S420 is a process of synchronizing devices to be used for communication with each other. That is, when the data conversion step (S140) is performed before the gateway transmission step (S150), between the POP-MCU, and when the data conversion step (S140) is performed before the server transmission step (S160) MCU Synchronize the server.

The transmission process S440 is a process of transmitting data to a receiver. That is, when the data conversion step (S140) is performed before the gateway transmission step (S150), it is a process of transmitting data to the MCU which is a receiver. And, if the data conversion step (S140) is performed before the server transmission step (S160) is a process of transmitting data to the server.

The receiving process S450 is a process of receiving data sent in the transmitting process S440. That is, the process of receiving data from a receiver such as an MCU or a server.

The packet checking process S470 is a process of checking whether there is an error, that is, data loss, in the data received by the receiver, and is a checksum checking process. A checksum is a form of redundancy check that, through error correction, protects the integrity of the transmitted data in space (electronic communication) or time (memory).

The retransmission request process (S480) is a process of requesting retransmission of data when there is a packet loss as a result of the check in the packet confirmation process (S470). In this case, the data may only request missing data. However, when the data capacity is not large, it is preferable to request the entire data again in a batch.

As described above, when the retransmission request process (S480) is performed, it is again performed from the transmission data conversion process (S430).

On the other hand, if there is no data lost in the packet checking process (S470), such information is stored and the data conversion step (S140) is completed.

The scope of the present invention is not limited to the above-exemplified embodiments, and many other modifications based on the present invention will be possible to those skilled in the art within the above technical scope.

S100. Product information input step S110. Production stage
S120. POP transfer step S130. Product recognition stage
S140. Data conversion step S150. Gateway transmission step
S160. Server transfer step S170. Determination
S180. Product count step S190. Inverse calculation step
S200. Set Product Count Step S210. Save step
S300. Learning phase

Claims (5)

A product information input step (S100) of inputting information on the produced product;
A production step of producing a product (S110);
A POP transmission step (S120) for transmitting the production related information input in the product information input step (S100) to a POP terminal;
A product recognition step (S130) of recognizing the product by the shape comparison of the product actually produced in the production step (S110);
A data conversion step (S140) of converting the result data of the POP transmission step (S120) and the product recognition step (S130) into data for heterogeneous networks;
A gateway transmission step (S150) or a server transmission step (S160) for transmitting the information passed through the data conversion step (S140) to a gateway or a server;
A determination step (S170) of comparing the production information by the POP transmission step (S120) with the product information recognized in the product recognition step (S130), and determining whether the information in both steps is identical to each other;
A result of the determination in step S170, when the information in the POP transmission step S120 and the product recognition step S130 coincide with each other, the production product count step S180 of increasing the inventory count of the corresponding product;
As a result of the determination in the determination step S170, when the information in the POP transmission step S120 and the product recognition step S130 coincide with each other, part information on a product produced by back ordering, etc. An inverse calculation step (S190) of inversely calculating a;
In addition to the reverse calculation step (S190), the set product count step (S200) for increasing the production inventory count of the POP set product;
Automatic product recognition and performance processing system comprising a; storing step (S210) for storing information about the production product. According to claim 1, The product recognition step (S130),
A photographing process of photographing an image of an actual product produced (S132);
A management process (S134) of managing the image photographed in the photographing process (S132);
Recognizing process (S136) to recognize the product by comparing the image taken in the photographing process (S132) with the learning information learned in the learning step; automatic product recognition and performance processing system comprising a. The method of claim 2, wherein the learning step compared in the recognition process (S136),
An item setting process (S310) for setting and managing items to be learned;
Shape learning process for learning the image of the product (S320);
Learning shape management process (S330) for managing the image learned in the shape learning process (S320);
Shape test process (S340) for testing the learned image;
A test shape management process (S350) for managing the shape tested by the shape test process (S340);
A recognition rate inspection process (S360) of checking a recognition rate level of the image tested by the test shape management process (S350);
A recognition rate determination step (S370) of determining whether the recognition rate level derived by the recognition rate checking process (S360) has a recognition rate of a predetermined level or more;
As a result of the determination in the recognition rate determination step (S370), if the recognition rate does not meet a predetermined level, the product automatic recognition and performance processing system, characterized in that the return to the shape learning process (S320). 4. The method according to any one of claims 1 to 3, wherein the gateway transmission step (S150) is optionally performed before the server transmission step (S160);
The data conversion step (S140),
Automatic product recognition and performance processing system, characterized in that performed before the gateway transmission step (S150) and server transmission step (S160), respectively. The method of claim 4, wherein the data conversion step (S140),
A method selection process (S410) of selecting a transmission method between heterogeneous types;
A synchronization process (S420) for synchronizing communication hardware;
A transmission data conversion process (S430) for converting transmission data according to a protocol;
A transmission process (S440) of transmitting data to the receiver;
Receiving process for receiving data at the receiver (S450);
A received data conversion process (S460) for converting the received data into received data according to a protocol;
A packet checking process (S470) for checking whether there is a packet loss in the received data;
As a result of the check in the packet checking step S470, if there is a packet loss, a retransmission request step S480 for requesting retransmission of data is included;
After the retransmission request process (S480), the transmission data conversion process (S430) is performed, characterized in that the automatic product recognition and performance processing system.

Images