KR102324352B1 - Ai-based program auto-generation system - Google Patents

Abstract

The present invention relates to a system for automatically generating an AI-based program that enables a component-based software to be developed by having a plurality of components disposed according to an input from a user to enable coding automation, and enables automatic distribution of a program. The system for automatically generating the AI-based program according to the present invention comprises: a user terminal; and a service server connected, to each other, to the user terminal through a communication network and comprising an editor part that automatically generates a program code based on an input from the user terminal, wherein the editor part comprises a component part comprising a plurality of components selected from the user terminal and disposed on a screen. Therefore, the present invention is capable of having an effect of developing a software even if not developed by a developer.

Description

AI-based program automatic generation system {AI-BASED PROGRAM AUTO-GENERATION SYSTEM}

The present invention relates to an AI-based program automatic generation system, and more particularly, by providing a plurality of components arranged according to an input from a user to enable coding automation, the development of component-based software is possible, and automatic distribution of the program is possible. It relates to a possible AI-based program automatic creation system.

Recently, companies are having difficulties in securing developers due to the increase in wages of software developers, and the secured developers are also experiencing difficulties in developing and constructing information systems due to lack of technical competency.

Meanwhile, information system development develops the information system desired by users through analysis, design, implementation, testing, and delivery procedures. Among them, the requirements analysis in the analysis stage and the program implementation in the implementation and test stage are important factors in information system development.

However, the current difficulties of developers due to the shortage of software developers and the resulting deterioration of information system quality are emerging as many problems. Accordingly, it is necessary to improve the quality of the information system and improve the developer shortage through a coding automation tool that enables non-developers to easily code software development.

In addition, applying the standardization guidelines (defense SHADE, public data standards) in development requires a lot of effort and a lot of work, and as a way to solve this, data standardization guidelines are applied when designing screens from the point of analysis of requirements through the use of coding automation tools. By complying with it, it is possible to comply with the software standardization policy by actively reflecting it in the standardization policy.

Republic of Korea Patent Publication No. 10-0962342 (Announced on June 10, 2010)

SUMMARY OF THE INVENTION An object of the present invention for solving the above-described problems is to provide an AI-based automatic program generation system capable of developing software even if not a developer.

Another object of the present invention is to provide an AI-based program automatic generation system capable of automatic code generation and automatic web service distribution and execution.

In order to achieve the above object, the AI-based program automatic generation system according to an embodiment of the present invention is connected to each other through a user terminal and a communication network with the user terminal, and generates a program code based on an input from the user terminal. and a service server including an editor unit that is automatically generated, wherein the editor unit includes a component unit including a plurality of components selected from the user terminal and arranged on a screen.

The editor unit further includes a component recommendation unit for recommending the component according to an input from the user terminal.

The input includes a sentence input, and the component recommendation unit recommends a component by analyzing a sentence according to the sentence input.

The input includes a voice input, and the component recommendation unit recommends a component by analyzing a voice according to the voice input.

The service server further includes an automatic web service distribution unit, and the automatic web service distribution unit supports an automatic DB generation unit that automatically creates a database when the component is arranged and an automatic distribution of the screen designed by the user terminal on the web. and an automatic distribution unit to, wherein the automatic DB generation unit generates the database by using the data standardization guidelines of public institutions announced by the Ministry of Public Administration and Security.

According to the AI-based program automatic generation system of the present invention, there is an effect that software can be developed even if you are not a developer.

According to the AI-based program automatic generation system of the present invention, there is an effect that automatic code generation and automatic web service distribution and execution are possible.

1 is a block diagram showing the configuration of an AI-based program automatic generation system according to the present invention.
2 is a block diagram showing the configuration of a service server according to the present invention.
3 is a view showing an initial screen provided by the service server according to the present invention.
4 is a diagram showing a state in which a database is automatically generated when a screen is designed by a component according to the present invention.
5 is a diagram showing a state in which the data standardization guidelines of public institutions announced by the Ministry of Public Administration and Security are utilized when the automatic DB generator according to the present invention creates a database.
6 is a diagram showing a state in which a screen designed using a component is automatically distributed.
7 is a flowchart showing the flow of services provided by the system according to the present invention.

Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. In adding reference numerals to components in each drawing, it should be noted that the same components are given the same reference numerals as much as possible even though they are indicated in different drawings.

And in describing the embodiment of the present invention, if it is determined that a detailed description of a related known configuration or function interferes with the understanding of the embodiment of the present invention, the detailed description thereof will be omitted.

In addition, in describing the components of the embodiment of the present invention, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only for distinguishing the components from other components, and the essence, order, or order of the components are not limited by the terms.

As used herein, the singular also includes the plural, unless the phrase specifically states otherwise. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components.

In this specification, software is used as a term including a program and data, and a program is used as a term including an application (application, app, app), web, and web page. Hereinafter, the program is described as a web. Program code includes source code.

In this specification, a program created by an operator or a user is referred to as a project.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of an AI-based automatic program generation system 1000 according to the present invention.

AI-based program automatic generation system 1000 according to the present invention, referring to FIG. 1 , a user terminal 200 and the user terminal 200 are connected to each other through a communication network, and input from the user terminal 200 is and a service server 100 including an editor unit 120 that automatically generates a program code based on the program code.

The user terminal 200 according to the present invention is capable of generating software including an application and/or a web page by accessing the service server 100 through a communication network. The user terminal 200 may design a screen on the web by arranging components to be described later.

The service server 100 includes an editor unit 120 . The editor unit 120 automatically generates program codes constituting software such as HTML, CSS, and JAVA SCRIPT based on the input from the user terminal 200 . The editor unit 120 may provide various design templates to create a web site based on an input from the user terminal 200 . The editor unit 120 supports the design of a screen on the web based on the components selected and/or arranged by the user terminal 200 .

The service server 100 automatically reflects the standardization of the database (hereinafter, 'DB') of the software configured through the program code generated by the editor unit 120, and supports automatic DB creation, automatic distribution, and automatic execution of the operating environment. It further includes an automatic web service distribution unit 130, and the detailed configuration of the automatic web service distribution unit 130 will be described later.

The configuration of the service server 100 according to the present invention will be described in more detail with reference to FIG. 2 as follows.

2 is a block diagram showing the configuration of the service server 100 according to the present invention.

The service server 100 according to the present invention, referring to FIG. 2 , a member management unit 110 , an editor unit 120 , an automatic web service distribution unit 130 , a my page unit 140 , and a customer support unit 150 ). includes

The member management unit 110 according to the present invention includes a login unit 111 , a simple login unit 112 , a password search unit 113 , an ID search unit 114 , and a member registration unit 115 .

The login unit 111 provides a function that the user terminal 200 can access to use the service server 100 according to the present invention. The simple login unit 112 provides a login function using APIs provided by Google, Kakao, Naver, and the like. The password search unit 113 provides a password search function through email authentication or mobile phone authentication from the user terminal 200 . The ID search unit 114 provides an ID search function through user information authentication or mobile phone authentication. The member registration unit 115 provides a function of registering member information in order for the user terminal 200 to use the service server 100 .

The editor unit 120 according to the present invention includes a main screen generation unit 121, a menu generation unit 122, a component unit 123, a component recommendation unit 124, a header and footer generation unit 125, and a pop-up screen. It includes a generation unit 126 , a preview unit 127 , a source view unit 128 , and a service start unit 129 .

The main screen generating unit 121 provides a function of generating the main screen of the project. The menu generating unit 122 provides a function of generating a menu of a project.

The component unit 123 includes a plurality of components selected from the user terminal 200 and arranged on the screen. The component unit 123 provides a plurality of different components configured for page production, such as TEXT, GRID, IMAGE, and BUTTON. The component unit 123 generates component source codes. The component unit 123 stores the generated component source code. The component unit 123 stores information on components selected and arranged by the user terminal 200 .

The component recommendation unit 124 provides a function of recommending specific components among a plurality of different components to the user terminal 200 according to an input from the user terminal 200 . The input from the user terminal 200 includes a click by an input device such as a mouse, a word or sentence input by an input device such as a keyboard, and a voice input by an input device such as a microphone. A detailed description of the component recommendation unit 124 will be described later.

The header and footer generation unit 125 provides a function of generating a header and footer of a project applied to all pages. The pop-up screen generating unit 126 provides a function of generating a pop-up that is opened through a button click or a separate action without moving to a menu. The preview unit 127 provides a function for previewing the produced project in the form of an actual web page. The source viewing unit 128 provides a function to inquire and modify the actual source of the produced project. The service start unit 129 provides a function of web hosting a project designed through the service server 100 in a serviceable form.

The automatic web service distribution unit 130 according to the present invention provides a web automation service. The automatic web service distribution unit 130 according to the present invention includes an automatic source code generation unit 131 , an automatic DB generation unit 132 , an automatic domain generation unit 133 , and an automatic distribution unit 134 .

The automatic source code generation unit 131 provides a function of automatically generating a source code so that a project designed using the service server 100 can be actually serviced and operated.

The automatic DB generation unit 132 provides a function of automatically generating tables and columns for data inquiry, input, modification, and deletion. In addition, the automatic DB generation unit 132 is configured by the user terminal 200 When placing components, a database is automatically created based on the components placed on the screen. On the other hand, the automatic DB generation unit 132 utilizes the data standardization guidelines of public institutions announced by the Ministry of Public Administration and Security when creating a database.

The automatic domain generation unit 133 provides a function of automatically generating a domain with the project name created by the user.

The automatic distribution unit 134 provides a function of distributing development sources to a web server so that the produced project can be serviced. The automatic distribution unit 134 supports automatic distribution on the web of the screen designed by the user terminal 200 according to the selection and arrangement of components.

The My Page unit 140 according to the present invention includes a project management dashboard unit 141 , a member information correction unit 142 , a purchase history confirmation unit 143 , and an inquiry history confirmation unit 144 .

The project management dashboard unit 141 provides a function of creating a new project or managing an existing project. The member information correction unit 142 provides a function for the user to modify his/her own information. The purchase history confirmation unit 143 provides a function to check the expiration date of purchase and payment information for the service provided by the service server 100 . The inquiry history confirmation unit 144 provides a function to check the inquiry message left in the customer center provided to the user terminal 200 by the service server 100 .

The customer support unit 150 according to the present invention includes an inquiry address unit 151 , a notice unit 152 , and an inquiry bulletin board unit 153 .

The inquiry unit 151 provides a function for checking information of a provider operating the service server 100 . The notice unit 152 provides a function to check the notice related to the service server 100 . The inquiry bulletin board 153 provides a function for registering an inquiry related to the service server 100 .

3 is a view showing an initial screen provided by the service server 100 according to the present invention.

The service server 100 provides a graphical user interface environment to the user terminal 200 . An initial screen in which the user terminal 200 accesses the service server 100 is shown in FIG. 3 . A plurality of components provided by the component unit 123 are provided on one side (A) of the initial screen provided by the service server 100 . The user terminal 200 may select any one of a plurality of components provided by the component unit 123 . The component selected by the user terminal 200 is placed on the screen. At this time, the arrangement position of the selected component can also be set (B). The user terminal 200 may design the screen of the app or web page by using the component. 3 shows a component capable of inputting and displaying product information, product inquiry, and product list as an example of a component.

4 is a view showing a state in which a database is automatically generated when a screen is designed by a component according to the present invention. It is a diagram showing the application of data standardization guidelines of public institutions.

The automatic DB generation unit 132 according to the present invention automatically creates a database when designing a screen by a component, referring to FIG. 4 . The user terminal 200 may design the screen of the app or web page by selecting, arranging and/or utilizing components as described above. The automatic DB generation unit 132 converts data input in individual components into a database, and at this time, it is possible to automatically reflect DB standardization.

The automatic DB generation unit 132 according to the present invention, with reference to FIG. 5 , designs a database by using the data standardization guidelines of public institutions announced by the Ministry of Public Administration and Security. In this specification, the guidelines for data standardization of public institutions announced by the Ministry of the Interior and Safety refer to the Ministry of Public Administration and Security, the Digital Government, the Information Standards and Guidelines Archive, and the establishment of database standardization guidelines for public institutions (Registration date: 2015.07.27. Author: Administrative Information Sharing Division) Data standardization guidelines according to the data standardization guidelines are posted at the following address.

‘https://www.mois.go.kr/frt/bbs/type001/commonSelectBoardArticle.do?bbsId=BBSMSTR_000000000045&nttId=46590’

Applying standardization guidelines (defense SHADE, public data standards) in development requires a lot of effort and man-hours, and as a solution for this, standardization policy by complying with data standardization guidelines when designing screens from the point of analysis of requirements through the use of coding automation tools By actively reflecting it in the software standardization policy, it can be met. Therefore, even if several people design and develop it, communication with each other is easy, it maintains consistency, and it is possible to develop a high-quality component-based information system with minimal effort.

6 is a diagram showing a state in which a screen designed using a component is automatically distributed.

The service server 100 according to the present invention includes an automatic web service distribution unit 130 that supports automatic distribution on the web of the screen designed by the user terminal 200 as described above. The project produced and designed by the editor unit 120 is web-hosted in a web serviceable form by the service start unit 129 .

The automatic web service distribution unit 130 is an automatic source code generation unit 131 that automatically generates a source code so that the service start unit 129 can actually service and operate a web-hosted project in a web serviceable form, a database An automatic DB generation unit 132 that automatically generates ), it is possible to automatically distribute the project including the screen designed by the user terminal 200 on the web.

7 is a flowchart showing the flow of services provided by the system according to the present invention.

Referring to FIG. 7 , the user terminal 200 accesses the service server 100 through a communication network, and joins the membership by the member registration unit 115 . The user terminal 200 registered as a member is logged in by the log-in unit 111 . The logged-in user terminal 200 is granted BASIC and Developer rights when purchasing the service, and Demo rights when not purchasing the service, depending on whether or not the service is purchased. The user terminal 200 generates a screen by using the component by the editor unit 120 and stores the generated screen. In addition, the user terminal 200 creates a project and stores the project. Saved projects can be downloaded as HTML files. The created project can automatically run the web service. The project created for the execution of the web service is built and automatically deployed.

As described above, the component unit 123 includes a plurality of different components. In this case, the component recommendation unit 124 may present a specific component to the user terminal 200 . The component recommendation unit 124 displays a specific component on one side A of the screen as shown in FIG. 3 , and the user terminal 200 can select a recommended component displayed on the screen.

The user terminal 200 includes an input device such as a mouse, a keyboard, and a microphone. The user terminal 200 may input a word, a sentence, or a voice through an input device in order to receive the component, and the user terminal 200 transmits the inputted word, sentence, or voice to the service server 100 . send The service server 100 that has received the word, sentence or voice from the user terminal 200 transmits the received word, sentence or voice to the component recommendation unit 124 . The component recommendation unit 124 may recommend a specific component based on the received word, sentence, or voice.

The service server 100 automatically generates a program code based on an input from the user terminal 200 . The input at this time includes a word input, a sentence input, and a voice input. The service server 100 provides a GUI and a user interface for input.

When the input from the user terminal 200 is a word input, the component recommendation unit 124 analyzes the word, calculates components with high relevance to the word, and displays them on the screen in order of high relevance.

When the input from the user terminal 200 is a sentence input, the component recommendation unit 124 analyzes the sentence, extracts a word, calculates a component highly related to the extracted word, and displays it on the screen in order of high correlation do.

When the input from the user terminal 200 is a voice input, the component recommendation unit 124 analyzes the voice, converts it into text (sentence form), extracts a word from the processed sentence, and correlates with the extracted word High components are calculated and displayed on the screen in the order of high relevance.

The component recommendation unit 124 may derive a user's pattern (action pattern) by analyzing the words, sentences, and voices input and received from the user terminal 200 to process words, sentences, and voices. 200), the component recommendation unit 124 analyzes text information for texts (including words or sentences) inputted through SNS applications such as text, KakaoTalk, Facebook, Twitter, etc. patterns can be derived. Meanwhile, when the component recommendation unit 124 receives a voice, a known technique for converting the received voice into text may be applied.

The component recommendation unit 124 performs natural language processing on the received words and sentences, recognizes words from the natural language-processed sentences, analyzes relationships between words in the recognized sentences, and assigns meanings to each word. can also interpret the text.

Through text interpretation in this way, the component recommendation unit 124 can more accurately understand the user's intention, and thereby can recommend a more suitable component to the user terminal 200 by accurately deriving the user's pattern (behavioral pattern). there will be

As an additional explanation for Natural Language Processing, natural language processing is a natural language (Natural Language) understanding or reconstruction of natural language that a computer can understand by mechanically analyzing human speech phenomena. It refers to various technologies expressed in a language that humans can understand.

Such natural language processing can be accomplished through morphological analysis and part-of-speech attachment. For morpheme analysis, for example, 'compound noun' can be divided into various methods such as 'complex + noun', 'complex + compound noun', 'compound + noun', etc., and the most appropriate decomposition result is selected among them. , a dynamic programming method called table parsing is used to select an appropriate result from among various analysis results. Specifically, if the probability of splitting into 'compound + noun' is greater than the probability of other analysis, it is an algorithm that selects it.

Part-of-speech attachment refers to attaching the most appropriate form of speech among the results obtained through morpheme analysis. A module, usually called a tagger, performs this function. This is a module that selects one analysis result suitable for the context from among various analysis results output by the morpheme analyzer. During analysis, an appropriate analysis result is selected using information that is a hint for disambiguation located on the left and right of the context. Usually, the tagger is implemented using a large-scale part-of-speech attachment, and the Hidden Markov Model (HMM) is widely used.

In the present invention, the text input from the user terminal 200 can be analyzed by the component recommendation unit 124 through the natural language processing described above.

In addition, when recognizing a sentence from the natural language-processed text, the component recommendation unit 124 refers to the time and place at which the text was input, and analyzes the relationship between the words in the recognized sentence to determine the meaning of each word. By giving the input, it is possible to more accurately grasp the user's intention and derive an appropriate analysis result by considering the time and place of the user's input.

In addition, in the present invention, a predetermined application is installed in the user terminal 200, and a recommended component suitable for the user's pattern is transmitted to the user terminal 200 based on the user's input pattern derived by the component recommendation unit 124, and the user It may be configured to be displayed on the screen of the terminal 200 .

On the other hand, the word derived by processing from the sentence or voice by the component recommendation unit 124 may be inserted into the web page in the form of a tag. In this case, the inserted tag may be used to facilitate the search when the web page is searched when the web page is later distributed.

When the input from the user terminal 200 is a voice, it is processed as described above and converted into text, and words are extracted from the processed text. Meanwhile, as an embodiment of the system according to the present invention, minutes may be automatically created in a meeting situation, a program may be produced based on the prepared minutes, or a specific component may be recommended based on the minutes.

In this case, the user terminals 200 may be a plurality of user terminals 200 (hereinafter, referred to as 'participant terminals') owned by the conference participants, respectively. The participant terminal includes a microphone for receiving the participant's voice. The participant terminal is connected to the service server 100 through a communication network. The service server 100 further includes a meeting minutes writing unit (not shown) for writing a voice recording file and/or meeting minutes.

The participant terminal detects the voice and voice levels input to each participant terminal, and when a voice level above or below a certain level is detected, it is automatically recognized that the owner of each participant terminal has started or finished speaking in the meeting.

An analog voice signal input by a microprocessor provided in the microphone of each participant's terminal is A/D converted and converted into voice data, and the microprocessor performs an operation based on the converted voice data to classify the inputted voice data made to do In addition, when a high voice level is instantaneously generated by the microprocessor provided in the microphone of the participant's terminal or there is no valid information, it is prevented that the owner of the terminal mistakenly believes that he has started speaking. On the other hand, a voice detection sensor configured to detect the voice level is included in the proper place of each participant's terminal.

In addition, when a high voice level is detected for some reason even though the owner does not speak at a predetermined location of each participant's terminal, a voice misinterpretation detection sensor is included to prevent the owner of the terminal from mistakenly saying that he has started speaking. .

Each participant terminal detects the voice level input by each meeting participant, and if the detected voice level is above a certain level, it is recognized that the owner of the participant terminal starts speaking in the meeting. Recognize that this is over.

The meeting minutes writing unit arranges the voice information sent by each participant terminal owned by these participants in chronological order according to the speech time information.

On the other hand, when two or more participants speak at the same time, a period of time in which the two voice data overlaps occurs, and when not speaking, a period of time in which there is no voice data may occur. The meeting minutes recorder integrates each participant's remarks into sorted voice information and creates a single voice transcript file.

In this way, the voice transcript file generated through the voice information extracted from the microphones of several dispersed attendee terminals is compared to the voice transcript file generated only with the voice information extracted through a single microphone installed in a specific area of the conference hall. You will be able to record and record clearly.

The audio is recorded in the audio transcript file generated by the meeting minutes recording unit. The component recommendation unit 124 may process the voice recorded in the voice transcript file as described above to convert it into text, analyze the text, and recommend a component to the user terminal 200 .

The editor unit 120 according to the present invention may further include an image input unit (not shown) for receiving an image from the user terminal 200 . The image input unit posts the received image on the web page. In this case, the image input unit can process an image using a frame dividing thread when posting an image.

In the proposed method, a frame-blocking method is used in which frames are divided into blocks and compared in order to reduce the size of image information. In this case, it is implemented using 16 threads for efficient data processing. Since all processed data is encrypted and stored by ARIA encryption algorithm, confidentiality of image information is guaranteed. Through the experimental results, it was confirmed that the image processing speed was improved by about 40% compared to the existing method using the original image, and the size of the image was reduced by about 60%.

The current image has been converted from the analog method of the past to the digital method, and a large amount of information can be stored. As such a large amount of information is stored, the cost of storage space has increased significantly, and countermeasures for this have become necessary. In addition, since it is a digital method, access to information is easy, and security measures are also required. A generally used image processing method is to take an image as it is, convert it into a digital image, and then store it. That is, it is a method of storing the image as it is without processing it in an appropriate way. However, since the existing methods allow forgery and falsification of images, many security-related problems such as invasion of privacy may occur. The present invention proposes a design and implementation of an image processing system using a frame division thread. In the proposed method, in order to reduce the size of image information, a frame block method is used, which divides and compares frames in blocks, and implements them using 16 threads. Since all processed data is encrypted and stored by ARIA encryption algorithm, confidentiality of image information is guaranteed.

Proposed image processing system

In the image processing process of the system proposed in the present invention, an image is received from the user terminal 200 and image processing and encryption are performed. The processed data is transmitted to the Viewer, and the original image is reproduced by performing decoding and image restoration in the Viewer. The image processing method in the present invention is a method using the correlation between frames, comparing the frame to be compared with the frame currently being reproduced, extracting and storing only meaningful pixel values. By using this method, frame capacity can be greatly reduced because only meaningful frames are extracted and combined into existing overlapping frames, excluding overlapping frames. Assuming that the video security system takes a picture with a camera and generates a reference frame every 60 seconds, the operation from shooting to transmission to the Viewer server is as shown in Figure 1.

<Figure 1. Flowchart of the proposed image processing>

As shown in Fig. 1, the proposed image processing process receives frames starting with camera shooting, finds the difference between the comparison target frame and the current frame, and blocks the values. At this time, the difference between frames and block are designed based on 16 split threads in consideration of efficiency. After that, the blocked frames are sorted and the first coordinates of each block (reference frame) are saved in a file. After that, using the ARIA encryption algorithm, encryption is performed on each frame data as a standard and sent to the Viewer, and the processed block alignment data is also sent to the Viewer.

16 split thread design

A moving picture is a technology that makes each still screen, that is, a frame, pass quickly to make it appear as if it is moving. Frame correlation refers to the difference between the two frames in that the background is the same but there is a moving object when successive frames are slightly different. In order to obtain a pixel difference using the frame correlation, a method of comparing the entire two frames in units of pixels has been mainly used in the prior art. However, since this method takes a lot of time for comparison, the present invention proposes a method of processing an image by dividing the image into 16 for more efficient frame comparison. A meaningful content is extracted by comparing the comparison target frame (reference frame) with the currently played frame. That is, in comparison with two frames, a fixed (reference) frame without a moving object is not extracted, but only the difference between the pixels of the two frames is obtained when a moving object is present. At this time, by comparing the pixel difference, overlapping data is removed, and only meaningful data is composed of variable blocks. In other words, after comparing the pixel difference, the same part is filled with white and only the moving object (the part different from the reference frame) is filled in the image. The reason for dividing a frame into blocks is to process meaningful pixel values quickly and accurately by objectifying them. In this case, blocks with no values are excluded from calculations because they are meaningless. Since both finding the difference between pixels (frames) and blocking frames can be performed after designating an area, the entire area can be divided into 16 and threads can be used at the same time. Therefore, the thread of each divided area performs the work of the next frame when the difference value between frames and frame block are performed at the same time period and the work for all areas is finished. By designing 16-split threads in this way, the image processing speed can be increased to prevent speed interference between the current frame and the next frame. Figure 2 shows an example of a frame 16-split thread configuration. A thread is the smallest program unit executed in the CPU and consists of information such as a thread ID, a set of registers, a stack, and a program counter. Since multi-threading enables memory sharing within the same process, efficient data processing is possible and system resource waste is reduced. In the present invention, when many threads are executed, the image is processed by dividing the data into 16 in consideration of the efficiency of data processing. In Figure 2, Locked is used to synchronize, and 16 threads belonging to one process share resources such as memory for parallel processing, so processing speed is very fast.

<Figure 2. Example of frame 16-split thread configuration>

frame reference

In the reference frame (image) designation, a reference image is generated every designated time by using the characteristics of an image to be photographed at a designated place. When a reference image is designated, subsequent frames are compared with this reference image and pixels. Assuming that the background brightness of the reference image and the subsequent frame is the same and there is a moving object, there is no pixel difference for the overlapping part with the previous one, so this part is filled with white, and only the moving object is extracted and filled in the image. In order to save the size of image files for comparison, a reference image generated every predetermined time is used. Figure 3 (a) shows the method for generating the reference image, and (b) shows the efficient algorithm for pixel comparison.

<Figure 3(a) Create a reference image>

<Figure 3(b) Pixel Comparison>

As shown in Figure 3(a) and Figure 3(b), the reference image is created every 60 seconds, and this file is encrypted and stored.

Frame Blocking and Block Alignment

When overlapping data is removed by comparing pixel differences, only meaningful data is composed of variable blocks. At this time, a portion with the same pixel is filled with white, only a portion different from the reference frame (image) is filled, and blocks are made to a certain size. If you want to extract more precise motion or data than this level, you can extract the data by lowering the reference value for pixel comparison or reducing the size of the block. After block formation, each block is filled in order from the top left. At this time, the coordinate values are put in front of each block so that the original position of each block can be found. Blocking and block sorting are performed by searching the entire image by block size. Only when the searched value is not white, the value is stored and blocked. Store the first coordinates of the block and sort the blocks. Figure 4 shows the blockization and block sorting processes that are executed simultaneously in 16 divisions.

<Figure 4. Image Blocking and Block Alignment Process>

When the image is processed through this process, the size of the image file can be reduced by up to 1/30 of the size of the original image file by reducing unnecessary data. When sorting, the first coordinates of each block are saved in the file, and they are read together when reading through the Viewer and used when rearranging blocks.

Image Viewer Design

In order to reproduce an image as a video, you need to know the processing speed (FPS, Frame Per Second) when taking each image to view the same video as the actual recorded video. If you do not know the FPS, the video may be too fast or too slow, and there may be parts missing during video playback. Therefore, when processing an image, the FPS must be saved in the image file. In the Viewer server, the processed image files are stored after shooting the video. At this time, since images are stored after blockization and block alignment, the viewer has to find the coordinates (coordinates for each reference frame) of each block (fragments) that have already been transmitted. Then, the original image is reproduced by connecting the blocks aligned to the reference image file according to the block coordinates.

Encryption of frame of reference

The proposed image processing system performs encryption and decryption based on the ARIA encryption algorithm. ARIA, a symmetric key algorithm, has a block length of 128 bits, and the length of a key of 128/192/256 bits can be variably used. Also, according to the size of the key, 10/12/14 rounds are repeated to execute encryption. The implemented encryption engine consists of a round module that performs encryption and decryption and a scheduler for generating key values. In the present invention, an encryption engine is designed and implemented using a key of 128 bits. The round function module consists of three main parts: key addition (AddRoundKey function), substitution layer (SubstLayer function), and diffusion layer (DiffLayer function). Key addition of the round function performs bitwise XOR operation on the 128-bit round key and the 128-bit input value. The permutation layer is composed of two types, and each layer has an 8-bit input and output S-box and an inverse transformation structure for them. The spreading layer performing the spreading function has a 16×16 involution structure and performs matrix operations.

Implementation result

In the present invention, the system was implemented by converting the existing ARIA C (C++) source code into C# code. Using each class, the definition part and the encryption and decryption part were reconstructed, and a fixed key value was set and used. Figure 5 shows the process of converting the C source code of the existing ARIA algorithm into C# code. Figure 6 shows an encrypted example of an aligned frame and a reference frame. Encryption is performed only on the last frame (aligned reference frame) that comes out after block alignment. As shown in Figure 6, if the aligned frame is encrypted, the encrypted frame (reference frame) is stored at the same address. At this time, the file to be encrypted is a reference frame file (bmp file). Due to the nature of this file, if the header part is damaged, the entire file becomes unreadable. Therefore, in the present invention, the bmp file, which is the header part of the frame, is partially encrypted to increase the encryption speed. Figure 7 shows the results of encryption (Ciphertext) and decryption (Decrypted) for three plaintext addresses of the aligned frame of Figure 6, respectively. As described above, the viewer in the video system reads the fixed frame file, the aligned file, and the coordinate file. Then, it reads the coordinates of the block from the coordinate file and applies these coordinates to the block alignment to move the block to its original position.

<Figure 5. Process of converting existing ARIA C code to C#>

<Figure 6. Example of aligned frame and encrypted frame>

<Figure 7. Encryption and Decryption Results for Aligned Frames>

Performance evaluation

Table 1 shows the experimental results of the existing original image and the proposed method while shooting an image for 5 minutes.

<Table 1. Comparison of image processing results and performance>

In the case of the original video, it is a method of shooting a video as it is without processing the video. In fact, it is the most used method in Korea and requires a lot of storage space, and anyone can access it, so there is a problem in data security. Proposed method 1 shows the result of performing the split-thread-based method proposed here, and Proposed method 2 shows the result of concurrently performing the split-thread-based and encryption method. As can be seen from Table 1, the FPS of Proposed Method 1 was improved by about 40% compared to the original video. And the capacity has also been reduced by about 75%, but anyone can access it, so it is exposed to security. In the case of Proposed Method 2, only the header part of the BMP file is partially encrypted and transmitted, so there is no difference from Method 1 in terms of speed. However, the average capacity increased to some extent through encryption, but security is excellent because access and confidentiality of unauthorized persons are guaranteed. Figure 8 shows a part of the screen executed in Viewer. The part shown in white in Figure 8 represents the part with no change (difference) from the reference frame. Among the four files, the image files stored as files in the server are block alignment (a) and reference frame (c). The reference frame and the block image (d) are images obtained by synthesizing the result of rearrangement of blocks in the reference frame, and the actual captured image is the same as (d). The encrypted image is decrypted only when the image is played back in the Viewer.

<Figure 8. Viewer execution screen>

The existing image processing does not process the image, but converts the original image into a digital image and uses it, so security is weak in forgery and falsification of the image. In the present invention, the implementation of a frame division thread-based CCTV security system using the ARIA algorithm has been described. The implemented system extracts only meaningful frames, excluding overlapping frames, and processes the images by synthesizing them with the existing overlapping frames, so that the frame capacity can be greatly reduced. The processing speed of the implemented security system was improved by about 40% compared to the existing image processing method, and the capacity was reduced by about 60%. In addition, the implemented system is effective in preventing infringement of personal information or privacy by a third party because the image data is encrypted and stored.

The user terminal 200 according to the present invention is a smart phone, a tablet personal computer (PC), a mobile phone, a video phone, and a desktop PC (desktoppersonal computer), which is equipped with software and can analyze data. , laptop personal computer (PC), netbook computer (netbook computer), personal digital assistant (PDA), portable multimedia player (PMP), wearable device (e.g. smart glasses, head-mounted device) -device (HMD, etc.) or a smart watch). The user terminal 200 may further include a communication unit for performing wired/wireless communication with the service server 100, and the communication unit may include BluetoothTM, Zigbee, Ultra Wide Band (UWB), Wireless USB, Near Field Communication (NFC), It may include electronic components for any one or more of the Wireless LANs.

The communication network according to the present invention refers to a connection structure capable of exchanging information between respective nodes such as terminals and servers. Examples of such a communication network include a 3rd Generation Partnership Project (3GPP) network, Long Term Evolution (LTE) ) network, 5G network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network) , a wifi network, a Bluetooth network, a satellite broadcasting network, an analog broadcasting network, a Digital Multimedia Broadcasting (DMB) network, and the like, but are not limited thereto.

A component according to the present invention means that a software module is made in advance, out of development by an existing coding method. The service server 100 may develop software by assembling components when developing a necessary application technology. By using component technology, a complex information system can be quickly built, and there is an advantage that it can be reused when constructing a similar information system. In particular, components are very important because it is necessary to make standardized software in the rapidly changing Internet environment.

A component refers to the division of the program's source code into modules and coding the parts to be changed dynamically. The source code of these components is stored for each program language, so it takes the form of an open structure to new platforms, languages, and versioned programs in the future. A component provided to the user in advance may be preset. Components selected by the user terminal 200 and/or arranged on the screen may be connected to each other according to their order, and a program may be generated. The service server 100 supports the user terminal 200 to connect to each other by dragging individual components on the screen.

Meanwhile, the component according to the present invention may be prepared and stored in advance, but is not limited thereto, and may be generated according to the needs of the operator of the service server 100 , the user of the user terminal 200 , and the like.

In the present specification, the member management unit 110, the editor unit 120, the automatic web service distribution unit 130, the my page unit 140, the customer support unit 150, the meeting minutes writing unit, the image input unit, and detailed configurations thereof are stored in the memory. They may be processors executing stored sequential execution processes. Alternatively, it may operate as software modules driven and controlled by a processor. Furthermore, the processor may be a hardware device.

For reference, the service provided by the service server 100 according to the present invention may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of computer readable media include hard disks, magnetic media such as floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floppy disks, and ROM, RAM, A hardware device specifically configured to store and execute program instructions, such as flash memory, may be included. Examples of program instructions include not only machine code such as generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

In this specification, a "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be implemented using two or more hardware, and two or more units may be implemented by one hardware.

The protection scope of the present invention is not limited to the description and expression of the embodiments explicitly described above. In addition, it is added once again that the protection scope of the present invention cannot be limited due to obvious changes or substitutions in the technical field to which the present invention pertains.

100: service server 110: member management unit
111: login unit 112: simple login unit
113: password retrieval unit 114: ID retrieval unit
115: member registration unit 120: editor unit
121: main screen generation unit 122: menu generation unit
123: component unit 124: component recommendation unit
125: Header and Footer generation unit 126: Pop-up screen generation unit
127: preview unit 128: source view unit
129: service start unit 130: automatic web service distribution unit
131: automatic source code generation unit 132: automatic DB generation unit
133: automatic domain generation unit 134: automatic distribution unit
140: my page unit 141: project management dashboard unit
142: Member information correction unit 143: Purchase history confirmation unit
144: inquiry history confirmation unit 150: customer support department
151: Inquiry Department 152: Notice Department
153: inquiry board 200: user terminal
1000: AI-based program automatic creation system

Claims (5)

user terminal; and
a service server connected to the user terminal and each other through a communication network, the service server including an editor unit for automatically generating a program code based on an input from the user terminal;
includes,
The editor unit,
a component unit including a plurality of components selected from the user terminal and arranged on a screen;
including,
The editor unit,
Further comprising a component recommendation unit for recommending the component according to the input from the user terminal,
The input includes a voice input,
The component recommendation unit,
Recommend a component by analyzing the voice according to the voice input,
The user terminal is provided with a plurality,
Each of the plurality of user terminals includes a microphone for receiving the voices of conference participants,
The service server further includes a meeting minutes writing unit for generating a voice transcript file through the voice information received from the plurality of distributed microphones,
The minutes of the meeting,
The plurality of voice information transmitted by the plurality of user terminals is arranged in the order of time according to the speaking time to generate the voice transcript file,
The component recommendation unit recommends a component by analyzing the voice included in the voice transcript file, and recommends the component based on the voice transcript file automatically created in a meeting situation. . delete The method according to claim 1,
The input further includes a sentence input,
The component recommendation unit,
AI-based program automatic generation system, characterized in that the component is recommended by analyzing the sentence according to the sentence input. delete 4. The method according to claim 1 or 3,
The service server is
It further includes an automatic web service distribution unit,
The automatic web service distribution unit,
an automatic DB generation unit that automatically creates a database when the components are placed; and
including; an automatic distribution unit that supports automatic distribution on the web of the screen designed by the user terminal;
The automatic DB generation unit, AI-based program automatic generation system, characterized in that for generating the database by using the data standardization guidelines of public institutions announced by the Ministry of Public Administration and Security.

Images