TW202129229A - Artificial intelligence patrol system and artificial intelligence patrol method - Google Patents

Artificial intelligence patrol system and artificial intelligence patrol method Download PDF

Info

Publication number
TW202129229A
TW202129229A TW109102386A TW109102386A TW202129229A TW 202129229 A TW202129229 A TW 202129229A TW 109102386 A TW109102386 A TW 109102386A TW 109102386 A TW109102386 A TW 109102386A TW 202129229 A TW202129229 A TW 202129229A
Authority
TW
Taiwan
Prior art keywords
information
image
portable device
artificial intelligence
inspection
Prior art date
Application number
TW109102386A
Other languages
Chinese (zh)
Other versions
TWI730605B (en
Inventor
陳文彬
Original Assignee
國眾電腦股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 國眾電腦股份有限公司 filed Critical 國眾電腦股份有限公司
Priority to TW109102386A priority Critical patent/TWI730605B/en
Application granted granted Critical
Publication of TWI730605B publication Critical patent/TWI730605B/en
Publication of TW202129229A publication Critical patent/TW202129229A/en

Links

Images

Landscapes

  • Testing And Monitoring For Control Systems (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

An artificial intelligence (AI) patrol system adapted to the 5th Generation entriprise private network comprises a portable system and an AI patrol analysis platform. The portable device comprises a camera module, a wireless communication module, a processor, and a notification output module. The camera module is configured to obtain an environment image. The wireless communicaton module sends an environment information associated with the environment image and receives a notification signal. The processor is configured to control the wireless communication module to send the environment information. The notification output module is configured to output the notification signal. The AI patrol analysis platform is configured to receive the environment information, load a three-dimensinal model database and a route information, perform a comparison procedure to generate a position information according to the environment information and the three-dimensional model database, generate and send the notification signal according to the position information and the route information.

Description

智慧巡檢系統及智慧巡檢方法Smart inspection system and smart inspection method

本發明涉及巡檢及影像辨識,特別是一種基於三維建模技術與高速通訊網路實現在石化業廠房內產生巡檢路線的系統及其方法。The invention relates to inspection and image recognition, in particular to a system and method for generating inspection routes in petrochemical industry factories based on three-dimensional modeling technology and high-speed communication network.

在石化業的廠房中具有複雜的管線、種類繁多的設備及儀表,為了維護廠房內各項資產設備的安全,安控中心需配置人員在廠房內按既定路線移動並且進行巡檢,巡檢人員除了例行性地監測設備指標數據,更需要隨時針對突發事件進行故障排除。In the petrochemical industry plant, there are complex pipelines, a wide variety of equipment and instruments. In order to maintain the safety of various assets and equipment in the plant, the security control center needs to deploy personnel to move in the plant according to the established route and conduct inspections. In addition to routinely monitoring equipment index data, it is also necessary to troubleshoot emergencies at any time.

然而,對於新進巡檢人員而言,恐無法立刻熟悉動線複雜的巡檢路線,而容易闖入具有高溫或高壓氣體的危險區域。需注意的是,傳統在廠區內設置定位點進行路線導引的方式並不適用於石化產業,這是由於若設置在定位點的電子感測器因故障產生火花,極有可能引起石化管線周邊外洩的可燃性氣體的連鎖爆炸反應,導致難以估計的損失。However, for new patrol personnel, they may not be familiar with the complicated patrol route immediately, and may easily break into dangerous areas with high temperature or high pressure gas. It should be noted that the traditional method of setting up positioning points in the factory area for route guidance is not suitable for the petrochemical industry. This is because if the electronic sensors installed at the positioning points produce sparks due to malfunctions, it is very likely to cause sparks around petrochemical pipelines. The chain explosion reaction of the leaked flammable gas has caused incalculable losses.

有鑑於此,本發明提出一種智慧巡檢系統及智慧巡檢方法,在第五代行動通訊技術(5th Generation mobile networks,5G)專用頻譜之企業專網布建而成的高速通訊環境底下實現石化業廠區內的智慧空間動線引導,以及實現針對每個巡檢人員的智慧派工。Accordingly, the present invention provides a system and wisdom wisdom patrol inspection method, implemented under the dedicated spectrum of the fifth generation of mobile communications technology (5 th Generation mobile networks, 5G ) corporate network deployment environment made of high-speed communication The intelligent space movement guidance in the petrochemical industry plant area, as well as the realization of the intelligent dispatch of labor for each inspector.

依據本發明一實施例敘述的一種智慧巡檢系統包括:可攜式裝置及人工智慧巡檢分析平台。可攜式裝置包括攝像模組、無線通訊模組、處理器及通知輸出模組。攝像模組用以取得環境影像。無線通訊模組用以發送關聯於環境影像之環境資訊及接收通知訊號。處理器電性連接攝像模組及無線通訊模組。處理器用以控制無線通訊模組發送環境資訊。通知輸出模組電性連接處理器。通知輸出模組用以輸出通知訊號。人工智慧巡檢分析平台通訊連接無線通訊模組以接收環境資訊。人工智慧巡檢分析平台用以載入三維模型資料庫及路線資訊。路線資訊包含三維模型資料庫的多個座標點的順序。人工智慧巡檢分析平台依據環境資訊及三維模型資料庫執行比對程序以獲取位置資訊。人工智慧巡檢分析平台依據位置資訊及路線資訊產生並發送通知訊號。A smart inspection system described according to an embodiment of the present invention includes a portable device and an artificial intelligence inspection and analysis platform. The portable device includes a camera module, a wireless communication module, a processor, and a notification output module. The camera module is used to obtain environmental images. The wireless communication module is used for sending environmental information related to the environmental image and receiving notification signals. The processor is electrically connected to the camera module and the wireless communication module. The processor is used for controlling the wireless communication module to send environmental information. The notification output module is electrically connected to the processor. The notification output module is used to output notification signals. The artificial intelligence inspection and analysis platform is connected to the wireless communication module to receive environmental information. The artificial intelligence inspection and analysis platform is used to load the 3D model database and route information. The route information includes the sequence of multiple coordinate points in the 3D model database. The artificial intelligence inspection and analysis platform executes a comparison procedure based on environmental information and a three-dimensional model database to obtain location information. The artificial intelligence inspection and analysis platform generates and sends notification signals based on location information and route information.

依據本發明一實施例敘述的一種智慧巡檢方法,包括:可攜式裝置之攝像模組取得環境影像;可攜式裝置之處理器控制可攜式裝置之無線通訊模組發送關聯於環境影像之環境資訊至人工智慧巡檢分析平台;人工智慧巡檢分析平台載入三維模型資料庫;人工智慧巡檢分析平台依據該環境資訊及三維模型資料庫執行比對程序以獲取位置資訊;人工智慧巡檢分析平台依據位置資訊及路線資訊產生並發送通知訊號至可攜式裝置,其中路線資訊包括三維模型資料庫的多個座標點的順序;可攜式裝置之無線通訊模組接收通知訊號;以及可攜式裝置之通知輸出模組輸出通知訊號。A smart inspection method described according to an embodiment of the present invention includes: a camera module of a portable device obtains an environmental image; a processor of the portable device controls a wireless communication module of the portable device to send an image associated with the environment The environmental information is sent to the artificial intelligence inspection and analysis platform; the artificial intelligence inspection and analysis platform loads the 3D model database; the artificial intelligence inspection and analysis platform executes a comparison procedure based on the environmental information and the 3D model database to obtain location information; artificial intelligence The patrol analysis platform generates and sends notification signals to the portable device based on location information and route information, where the route information includes the sequence of multiple coordinate points in the 3D model database; the wireless communication module of the portable device receives the notification signal; And the notification output module of the portable device outputs the notification signal.

以上之關於本揭露內容之說明及以下之實施方式之說明係用以示範與解釋本發明之精神與原理,並且提供本發明之專利申請範圍更進一步之解釋。The above description of the disclosure and the following description of the embodiments are used to demonstrate and explain the spirit and principle of the present invention, and to provide a further explanation of the scope of the patent application of the present invention.

以下在實施方式中詳細敘述本發明之詳細特徵以及優點,其內容足以使任何熟習相關技藝者了解本發明之技術內容並據以實施,且根據本說明書所揭露之內容、申請專利範圍及圖式,任何熟習相關技藝者可輕易地理解本發明相關之目的及優點。以下之實施例係進一步詳細說明本發明之觀點,但非以任何觀點限制本發明之範疇。The detailed features and advantages of the present invention will be described in detail in the following embodiments. The content is sufficient to enable anyone familiar with the relevant art to understand the technical content of the present invention and implement it accordingly, and according to the content disclosed in this specification, the scope of patent application and the drawings. Anyone who is familiar with relevant skills can easily understand the purpose and advantages of the present invention. The following examples further illustrate the viewpoints of the present invention in detail, but do not limit the scope of the present invention by any viewpoint.

請參考圖1,其繪示本發明一實施例敘述的智慧巡檢系統100的方塊架構圖。所述的智慧巡檢系統100適用於已佈建高速行動通訊網路N的石化業廠房。在一實施例中,高速行動通訊網路N由多個無線存取點AP及基地台所組成,這些無線存取點AP係透過第五代行動通訊技術(5th Generation mobile networks,5G)專用頻譜之企業專網彼此互相連接。Please refer to FIG. 1, which shows a block diagram of a smart inspection system 100 described in an embodiment of the present invention. The smart inspection system 100 is applicable to petrochemical plants where a high-speed mobile communication network N has been deployed. In one embodiment, the high speed mobile communication network N by the radio access point AP and a plurality of base stations are composed of the wireless access point AP through the fifth-generation mobile communication system technologies (5 th Generation mobile networks, 5G ) dedicated spectrum of Enterprise private networks are connected to each other.

如圖1所示,智慧巡檢系統100包括:可攜式裝置3及人工智慧巡檢分析平台5。可攜式裝置3係透過前述無線存取點AP其中一者通訊連接至人工智慧巡檢分析平台5。As shown in FIG. 1, the smart inspection system 100 includes a portable device 3 and an artificial intelligence inspection and analysis platform 5. The portable device 3 is connected to the artificial intelligence inspection and analysis platform 5 through one of the aforementioned wireless access points AP.

如圖1所示,可攜式裝置3包括攝像模組32、無線通訊模組34、處理器36及通知輸出模組38。處理器36電性連接至攝像模組32、無線通訊模組34及通知輸出模組38。人工智慧巡檢分析平台5通訊連接無線通訊模組34以接收環境資訊。在一實施例中,人工智慧巡檢分析平台5係一伺服器或伺服器叢集,並且通訊連接至以5G實現的高速行動通訊網路N。藉此,人工智慧巡檢分析平台5可在較小延遲的條件下接收到每個可攜式裝置3回傳的環境資訊。As shown in FIG. 1, the portable device 3 includes a camera module 32, a wireless communication module 34, a processor 36 and a notification output module 38. The processor 36 is electrically connected to the camera module 32, the wireless communication module 34 and the notification output module 38. The artificial intelligence inspection and analysis platform 5 is communicatively connected to the wireless communication module 34 to receive environmental information. In one embodiment, the artificial intelligence inspection and analysis platform 5 is a server or a cluster of servers, and is communicatively connected to a high-speed mobile communication network N implemented by 5G. In this way, the artificial intelligence inspection and analysis platform 5 can receive the environmental information returned by each portable device 3 with a relatively small delay.

請參考圖2,其繪示本發明一實施例敘述的可攜式裝置3的外觀示意圖。在一實施例中,可攜式裝置3為供巡檢人員配戴的一智慧型頭盔。因攝像模組32之設置鄰近於配戴智慧型頭盔型態的可攜式裝置3的巡檢人員的眼部位置,故巡檢人員可輕易地透過攝像模組32收集其所看到的廠房內部的影像資訊。在其他實施例中,可攜式裝置3的外觀型態亦可以是智慧型眼鏡、智慧型手機、平板電腦或個人數位助理(Personal Digital Assistant,PDA)等。本發明對於可攜式裝置3的外觀或硬體類型並未加以限制。Please refer to FIG. 2, which shows a schematic diagram of the appearance of the portable device 3 described in an embodiment of the present invention. In one embodiment, the portable device 3 is a smart helmet worn by inspectors. Since the camera module 32 is set close to the eye position of the inspector wearing the portable device 3 of the smart helmet type, the inspector can easily collect the factory buildings he sees through the camera module 32 Internal image information. In other embodiments, the appearance of the portable device 3 may also be smart glasses, smart phones, tablet computers, or personal digital assistants (Personal Digital Assistant, PDA), etc. The present invention does not limit the appearance of the portable device 3 or the type of hardware.

請參考圖3,其繪示本發明一實施例敘述的智慧巡檢方法的流程圖。所述的智慧巡檢方法適用於前述的智慧巡檢系統100。以下結合圖3繪示的智慧巡檢方法的流程,進一步說明智慧巡檢系統100中各項元件的組成及實現方式。Please refer to FIG. 3, which shows a flowchart of a smart inspection method described in an embodiment of the present invention. The smart inspection method described above is applicable to the aforementioned smart inspection system 100. The following describes the composition and implementation of various components in the smart inspection system 100 in conjunction with the flow of the smart inspection method shown in FIG. 3.

請參考步驟S1,攝像模組32取得環境影像。詳言之,在石化業廠房中,在巡檢人員巡檢時,其所配戴的可攜式裝置3中的攝像模組32已被啟動,且攝像模組32可自動拍攝當前位置的環境影像。舉例來說,攝像模組32係高畫質攝影機,藉此拍攝高解析度的石化廠內部環境的影像。在一實施例中,攝像模組32包含二個以上的鏡頭,其中一個鏡頭設置於額頭的位置,另一個鏡頭設置於頭頂的位置。因此攝像模組32可獲取不同視角的環境影像。Please refer to step S1, the camera module 32 obtains an environment image. In detail, in the petrochemical industry plant, during the inspection by the inspectors, the camera module 32 in the portable device 3 worn by the inspector has been activated, and the camera module 32 can automatically capture the environment at the current location image. For example, the camera module 32 is a high-definition camera for shooting high-resolution images of the internal environment of the petrochemical plant. In one embodiment, the camera module 32 includes more than two lenses, one of which is set on the forehead and the other on the top of the head. Therefore, the camera module 32 can obtain environmental images with different viewing angles.

在一實施例中,在步驟S1之前(即:可攜式裝置3之攝像模組32取得環境影像之前,相當於巡檢人員開始巡檢之前),更包括下列步驟:可攜式裝置3之攝像模組32取得起點影像,在取得起點影像之後,處理器36控制無線通訊模組34發送起點影像及復歸訊號;以及人工智慧巡檢分析平台5依據接收到的復歸訊號及起點影像執行比對程序以獲得確認訊號,並發送確認訊號至可攜式裝置3。詳言之,當巡檢人員開始巡檢之前,需先到達位於廠房內的巡檢起點。在巡檢人員到達起點後,巡檢人員啟動可攜式裝置3且攝像模組32拍攝起點影像,然後巡檢人員對可攜式裝置3進行特定的操作(例如在攝像模組32的鏡頭前比出特定的手勢供處理器36進行影像辨識)以發送復歸訊號,這代表巡檢人員通知安控中心其已就定位。在安控中心的人工智慧巡檢分析平台5收到復歸訊號之後,依據三維模型資料庫中作為快取的多個預設影像(巡檢起點的預設影像)執行比對程序(或是直接依據三維模型資料庫的所有影像資料進行比對程序),藉此確認可攜式裝置3所回傳的起點影像是否正確,若正確,則發送確認訊號給可攜式裝置3。完成上述的起點就定位的程序之後,巡檢人員才可開始既定路線的巡檢並且拍攝巡檢路線上的環境影像。In one embodiment, before step S1 (that is, before the camera module 32 of the portable device 3 obtains environmental images, which is equivalent to before the inspector starts the inspection), the following steps are further included: The camera module 32 obtains the starting point image. After obtaining the starting point image, the processor 36 controls the wireless communication module 34 to send the starting point image and the reset signal; and the artificial intelligence inspection and analysis platform 5 performs comparison based on the received reset signal and the starting point image The program obtains the confirmation signal and sends the confirmation signal to the portable device 3. In detail, before the inspection personnel start the inspection, they must first arrive at the starting point of the inspection in the factory. After the inspector arrives at the starting point, the inspector activates the portable device 3 and the camera module 32 captures an image of the starting point, and then the inspector performs specific operations on the portable device 3 (for example, in front of the lens of the camera module 32). A specific gesture is compared for the processor 36 to perform image recognition) to send a reset signal, which means that the inspector informs the security control center that it has been located. After the artificial intelligence inspection and analysis platform 5 of the security control center receives the reset signal, it executes the comparison program (or directly Perform a comparison procedure based on all the image data in the 3D model database) to confirm whether the starting image returned by the portable device 3 is correct, and if it is correct, a confirmation signal is sent to the portable device 3. After completing the above-mentioned procedure of locating at the starting point, the patrol personnel can start the patrol of the established route and shoot the environmental images on the patrol route.

在另一實施例中,在步驟S1之前,更包括下列步驟:攝像模組32取得起點影像,處理器36比對起點影像及本身儲存的多個預設影像,當起點影像與這些預設影像之中一者相符,處理器36控制無線通訊模組34發送對應於相符的預設影像的預設座標點。一般而言,巡檢人員在開始巡檢之前,通常係於廠房內的固定地點集合後再出發。因此,處理器36中預先儲存多個巡檢起始點周圍的環境影像作為預設影像,且每個預設影像對應於一個預設座標點。因此,在處理器36比對起點影像與預設影像並確認兩者相符之後,便可發送本次巡檢人員起始位置的座標點至位於安控中心的人工智慧巡檢分析平台5,且在無線通訊模組34發送預設座標點之後,始以攝像模組32取得環境影像。透過上述機制,本發明一實施例的智慧巡檢系統100可完整掌握每一個巡檢人員是否從規定位置開始巡檢,而且,可彈性地決定每次巡檢人員的巡檢路線,例如:要求巡檢人員移動到正確的巡檢起始點,或是依據當前的起始點調整巡檢人員接下來的巡檢路線。In another embodiment, before step S1, it further includes the following steps: the camera module 32 obtains the starting point image, and the processor 36 compares the starting point image with a plurality of preset images stored by itself, when the starting point image and the preset images If one of them matches, the processor 36 controls the wireless communication module 34 to send a preset coordinate point corresponding to the matched preset image. Generally speaking, the inspection personnel usually gather at a fixed place in the factory before starting the inspection. Therefore, the processor 36 pre-stores a plurality of environmental images around the starting point of the inspection as preset images, and each preset image corresponds to a preset coordinate point. Therefore, after the processor 36 compares the starting point image with the preset image and confirms that the two are consistent, the coordinate point of the starting position of the patrol personnel can be sent to the artificial intelligence patrol analysis platform 5 in the security control center, and After the wireless communication module 34 sends the preset coordinate points, the camera module 32 starts to obtain the environment image. Through the above mechanism, the smart inspection system 100 of an embodiment of the present invention can fully grasp whether each inspector starts the inspection from a specified position, and can flexibly determine the inspection route of each inspector, for example: The inspector moves to the correct starting point of the inspection, or adjusts the next inspection route of the inspector according to the current starting point.

請參考步驟S2,處理器36控制無線通訊模組34發送環境資訊。環境資訊關聯於環境影像。在一實施例中,環境資訊即係環境影像。在另一實施例中,處理器36預先對攝像模組32拍攝到的環境影像進行影像前處理程序,藉此減少後續的環境資訊傳輸所佔用的資料頻寬。關於影像前處理程序,舉例來說可包含以下兩個步驟:首先建立與環境影像中的物件姿態無關的物件特徵,例如處理器36將環境影像旋轉,並使用局部座標對管線或設備的表面影像點進行編碼;其次在三維場景中偵測感興趣區域(Region of Interest,ROI),例如處理器36採用三維線索法(3-D cueing)在環境影像中偵測感興趣區域。最後處理器36送出的環境資訊為感興趣區域的影像。在具有多個配戴可攜式裝置3的巡檢人員同時巡檢時,上述機制除了可縮短影像傳輸到安控中心的時間,更可提升智慧巡檢系統100整體運作的效率。Please refer to step S2, the processor 36 controls the wireless communication module 34 to send environmental information. Environmental information is associated with environmental images. In one embodiment, the environmental information is environmental images. In another embodiment, the processor 36 pre-processes the environmental images captured by the camera module 32 to reduce the data bandwidth occupied by the subsequent environmental information transmission. Regarding the image pre-processing procedure, for example, it may include the following two steps: first, create object features that are not related to the posture of the object in the environment image. For example, the processor 36 rotates the environment image and uses local coordinates to image the surface of the pipeline or equipment. The points are encoded; secondly, a region of interest (Region of Interest, ROI) is detected in the three-dimensional scene. For example, the processor 36 uses a three-dimensional cueing method (3-D cueing) to detect the region of interest in the environmental image. Finally, the environmental information sent by the processor 36 is an image of the region of interest. When multiple inspectors wearing portable devices 3 perform simultaneous inspections, the above-mentioned mechanism can not only shorten the time for image transmission to the security control center, but also improve the overall operation efficiency of the smart inspection system 100.

請參考步驟S3,人工智慧巡檢分析平台5載入三維模型資料庫並接收環境資訊。整體而言,人工智慧巡檢分析平台5可載入三維模型資料庫及路線資訊,並依據環境資訊及三維模型資料庫執行比對程序以獲取位置資訊,以及依據位置資訊及路線資訊產生並發送通知訊號。在一實施例中,所述三維模型資料庫包括一或多個電腦輔助設計(Computer Aided Design,CAD)檔案及多個座標點。舉例來說,可取用石化業廠房在設計時的CAD檔建立此三維模型資料庫,所述的多個座標點包括巡迴路線的起點、終點及中繼點。請參考圖4,其係繪示巡檢路線上的多個座標點a,b及c的示意圖。在其他實施例中,可運用雷射掃描或石化廠詳細施工圖來逆向建立管路與其他設備的三維模型資料庫。舉例來說,利用三維掃瞄儀器取得管路或設備表面的空間資訊,並利用點雲辨識軟體將其轉換為具有實體意義的立體模型。舉另一例來說,根據詳細施工圖找出管線的起迄點或參考點的座標,然後藉由CAD軟體系統半自動地建立三維模型資料庫。Please refer to step S3, the artificial intelligence inspection and analysis platform 5 loads the 3D model database and receives environmental information. On the whole, the artificial intelligence inspection and analysis platform 5 can load a 3D model database and route information, and perform a comparison procedure based on the environmental information and the 3D model database to obtain location information, and generate and send based on the location information and route information Notification signal. In one embodiment, the 3D model database includes one or more Computer Aided Design (CAD) files and a plurality of coordinate points. For example, the CAD file of the petrochemical industry plant during the design can be used to create this three-dimensional model database, and the multiple coordinate points include the starting point, the end point and the relay point of the patrol route. Please refer to FIG. 4, which is a schematic diagram showing multiple coordinate points a, b, and c on the inspection route. In other embodiments, laser scanning or detailed construction drawings of a petrochemical plant can be used to reversely build a three-dimensional model database of pipelines and other equipment. For example, use a three-dimensional scanning instrument to obtain the spatial information of the pipeline or equipment surface, and use point cloud recognition software to convert it into a solid model with physical meaning. For another example, find out the coordinates of the start and end points or reference points of the pipeline according to the detailed construction drawings, and then semi-automatically build a 3D model database by the CAD software system.

請參考步驟S4,人工智慧巡檢分析平台5執行比對程序產生位置資訊。詳言之,人工智慧巡檢分析平台5依據環境資訊比對三維模型資料庫的多個座標點所對應的樣本影像並判斷其中是否具有相符者。在一實施例中,環境資訊為環境影像或感興趣區域的影像,且比對程序係比對二維影像的相似程度。在另一實施例中,攝像模組32中更包含深度攝影機或是距離感測器,並且在可攜式裝置3回傳的環境資訊中,更包含以深度攝影機或距離感測器所取得的影像深度資訊,且人工智慧巡檢分析平台5所執行的比對程序係比對三維影像的相似程度。在又一實施例中,人工智慧巡檢分析平台5接收可攜式裝置3在同一時間拍攝到的多個視角的環境影像(透過可攜式裝置3上多個不同設置位置的攝影模組32的鏡頭)並且分別執行比對程序,藉此確認這些環境影像是否屬於同一個位置的座標點範圍。透過上述機制,本發明提出的智慧巡檢系統100可提高以環境影像辨識巡檢人員位置的精準度。另需補充的是,人工智慧巡檢分析平台5可同時針對不同硬體架構的可攜式裝置3,選擇性地執行二維影像的比對程序或是三維影像的辨識程序,本發明對此不予限制。Please refer to step S4, the artificial intelligence inspection and analysis platform 5 executes a comparison procedure to generate location information. In detail, the artificial intelligence inspection and analysis platform 5 compares the sample images corresponding to multiple coordinate points in the three-dimensional model database according to the environmental information and determines whether there is a match among them. In one embodiment, the environment information is an environment image or an image of a region of interest, and the comparison procedure is to compare the similarity of the two-dimensional image. In another embodiment, the camera module 32 further includes a depth camera or a distance sensor, and the environmental information returned by the portable device 3 further includes information obtained by the depth camera or the distance sensor. The image depth information, and the comparison program executed by the artificial intelligence inspection and analysis platform 5 compares the similarity of the three-dimensional image. In another embodiment, the artificial intelligence inspection and analysis platform 5 receives environmental images of multiple viewing angles captured by the portable device 3 at the same time (through multiple camera modules 32 in different positions on the portable device 3). Lens) and perform the comparison procedure respectively to confirm whether these environmental images belong to the coordinate point range of the same position. Through the above-mentioned mechanism, the smart inspection system 100 proposed in the present invention can improve the accuracy of identifying the position of the inspector based on the environmental image. It should also be added that the artificial intelligence inspection and analysis platform 5 can selectively execute two-dimensional image comparison procedures or three-dimensional image recognition procedures for portable devices 3 with different hardware architectures at the same time. No restrictions.

請參考步驟S5,人工智慧巡檢分析平台5依據位置資訊及路線資訊產生並發送通知訊號。路線資訊包含三維模型資料庫的多個座標點的順序。舉例來說,如圖4所示,一個可行的路線資訊如圖4中的座標點a、b及c所構成的巡檢路線。實務上,人工智慧巡檢分析平台5依據巡檢人員當前所處的位置,發送到達下一個座標點的相關通知訊號。舉例來說,通知內容可包括下一次轉彎的方向以及在下一次轉彎之前預計前進的距離。Please refer to step S5. The artificial intelligence inspection and analysis platform 5 generates and sends a notification signal based on the location information and route information. The route information includes the sequence of multiple coordinate points in the 3D model database. For example, as shown in FIG. 4, a feasible route information is a patrol route composed of coordinate points a, b, and c in FIG. In practice, the artificial intelligence inspection and analysis platform 5 sends relevant notification signals to reach the next coordinate point based on the current position of the inspector. For example, the notification content may include the direction of the next turn and the estimated distance to travel before the next turn.

請參考步驟S6,無線通訊模組34接收通知訊號且通知輸出模組38輸出通知訊號。無線通訊模組34除了可發送關聯於環境影像之環境資訊外,更用以接收通知訊號。請參考圖1,通知輸出模組38電性連接處理器36。通知輸出模組38用以輸出通知訊號。在一實施例中,通知訊號係包含一路線影像,且通知輸出模組38係一顯示裝置,例如圖2所示的智慧頭盔上的眼鏡鏡片。顯示裝置可呈現路線影像。因此巡檢人員可按照路線影像指示的方向逐步完成巡檢路線。在另一實施例中,通知訊號係包含一語音,且通知輸出模組38係一揚聲器,該揚聲器用以播放該語音。因此巡檢人員可按照揚聲器指示的語音逐步完成巡檢路線。Please refer to step S6, the wireless communication module 34 receives the notification signal and the notification output module 38 outputs the notification signal. The wireless communication module 34 can not only send environmental information related to the environmental image, but also receive notification signals. Please refer to FIG. 1, the notification output module 38 is electrically connected to the processor 36. The notification output module 38 is used for outputting notification signals. In one embodiment, the notification signal includes a route image, and the notification output module 38 is a display device, such as the glasses lens on the smart helmet shown in FIG. 2. The display device can present route images. Therefore, the inspection personnel can gradually complete the inspection route according to the direction indicated by the route image. In another embodiment, the notification signal includes a voice, and the notification output module 38 is a speaker, and the speaker is used to play the voice. Therefore, the inspection personnel can gradually complete the inspection route according to the voice indicated by the speaker.

在又一實施例中,可攜式裝置3更包括一慣性測量模組(Inertial Measurement Unit,IMU)電性連接處理器36。慣性測量模組用以取得可攜式裝置3之移動資訊。且前述的環境影像係一第一環境影像,環境資訊係一第一環境資訊,位置資訊係一第一位置資訊。在步驟S1之後,處理器36更用以控制攝像模組32在取得第一環境影像並經過一間隔時間之後取得第二環境影像。無線通訊模組34更用以發送關聯於第二環境影像的第二環境資訊及移動資訊。人工智慧巡檢分析平台5更用以接收第二環境資訊,依據第二環境資訊、三維模型資料庫、間隔時間及移動資訊執行比對程序以獲得一第二位置資訊。並依據第一位置資訊、第二位置資訊及路線資訊選擇性地產生一警示訊號。實務上,為避免新進的巡檢人員在複雜的廠房中迷路,因此攝像模組32每隔一段時間即回傳當下的環境影像供人工智慧巡檢分析平台5分析巡檢人員是否按既定巡檢路線前進。若巡檢人員脫離預定的巡檢路線,則人工智慧巡檢分析平台5將發送警示訊號,可攜式裝置3在接收到警示訊號後,將透過通知輸出模組38告知巡檢人員,例如播放「已進入危險區域,請儘速退後」的語音,或是在顯示裝置上呈現具有明顯提醒作用的圖層或文字訊息。In another embodiment, the portable device 3 further includes an Inertial Measurement Unit (IMU) electrically connected to the processor 36. The inertial measurement module is used to obtain the mobile information of the portable device 3. And the aforementioned environmental image is a first environmental image, the environmental information is a first environmental information, and the location information is a first location information. After step S1, the processor 36 is further used to control the camera module 32 to obtain the first environment image and obtain the second environment image after an interval has elapsed. The wireless communication module 34 is further used for sending second environment information and mobile information related to the second environment image. The artificial intelligence inspection and analysis platform 5 is further used to receive the second environment information, and perform a comparison process based on the second environment information, the three-dimensional model database, the interval time, and the movement information to obtain a second location information. And selectively generate a warning signal based on the first location information, the second location information, and the route information. In practice, in order to prevent new patrol personnel from getting lost in the complex plant, the camera module 32 will return the current environmental images at regular intervals for the artificial intelligence patrol analysis platform 5 to analyze whether the patrol personnel are in accordance with the scheduled inspection The route goes forward. If the inspector departs from the scheduled inspection route, the artificial intelligence inspection analysis platform 5 will send a warning signal, and the portable device 3 will notify the inspector through the notification output module 38 after receiving the warning signal, such as playing "You have entered a dangerous area, please go back as soon as possible" voice, or display a layer or text message with obvious reminder function on the display device.

綜上所述,本發明提出的智慧巡檢系統可提升巡檢執行效能,提高工安安全性。對場域進行人員管制追蹤,並進行巡檢資訊收集回饋,及石化廠房內資產設備的監控管理。In summary, the smart inspection system proposed by the present invention can improve the execution efficiency of the inspection and improve the safety of industrial safety. Carry out personnel control tracking on the field, and carry out inspection information collection and feedback, and the monitoring and management of assets and equipment in the petrochemical plant.

雖然本發明以前述之實施例揭露如上,然其並非用以限定本發明。在不脫離本發明之精神和範圍內,所為之更動與潤飾,均屬本發明之專利保護範圍。關於本發明所界定之保護範圍請參考所附之申請專利範圍。Although the present invention is disclosed in the foregoing embodiments, it is not intended to limit the present invention. All changes and modifications made without departing from the spirit and scope of the present invention fall within the scope of the patent protection of the present invention. For the scope of protection defined by the present invention, please refer to the attached scope of patent application.

100:智慧巡檢系統 N:高速行動通訊網路 AP:無線存取點 3:可攜式裝置 32:攝像模組 34:無線通訊模組 36:處理器 38:通知輸出模組 5:人工智慧巡檢分析平台 a、b、c:巡檢路線座標點 S1~S6:步驟100: Smart inspection system N: High-speed mobile communication network AP: wireless access point 3: portable device 32: camera module 34: wireless communication module 36: processor 38: Notification output module 5: Artificial intelligence inspection and analysis platform a, b, c: inspection route coordinate points S1~S6: steps

圖1係依據本發明一實施例繪示的智慧巡檢系統的方塊架構圖。 圖2係依據本發明一實施例繪示的可攜式裝置的外觀示意圖。 圖3係依據本發明一實施例繪示的智慧巡檢方法的流程圖。 圖4係繪示本發明一實施例的巡檢路線的示意圖。FIG. 1 is a block diagram of a smart inspection system according to an embodiment of the present invention. FIG. 2 is a schematic diagram showing the appearance of a portable device according to an embodiment of the present invention. FIG. 3 is a flowchart of a smart inspection method according to an embodiment of the present invention. Fig. 4 is a schematic diagram showing a patrol route according to an embodiment of the present invention.

100:智慧巡檢系統100: Smart inspection system

N:高速行動通訊網路N: High-speed mobile communication network

AP:無線存取點AP: wireless access point

3:可攜式裝置3: portable device

32:攝像模組32: camera module

34:無線通訊模組34: wireless communication module

36:處理器36: processor

38:通知輸出模組38: Notification output module

5:人工智慧巡檢分析平台5: Artificial intelligence inspection and analysis platform

Claims (10)

一種智慧巡檢系統,所述的智慧巡檢系統包括:一可攜式裝置,包括:一攝像模組,用以取得一環境影像;一無線通訊模組,該無線通訊模組用以發送關聯於該環境影像之一環境資訊及接收一通知訊號;一處理器,電性連接該攝像模組及該無線通訊模組,該處理器用以控制該無線通訊模組發送該環境資訊;及一通知輸出模組,電性連接該處理器,該通知輸出模組用以輸出該通知訊號;以及一人工智慧巡檢分析平台,通訊連接該無線通訊模組以接收該環境資訊,該人工智慧巡檢分析平台用以載入一三維模型資料庫及一路線資訊,該路線資訊包括該三維模型資料庫的多個座標點的一順序,該人工智慧巡檢分析平台依據該環境資訊及該三維模型資料庫執行一比對程序以獲取一位置資訊,並依據該位置資訊及該路線資訊產生並發送該通知訊號。A smart patrol inspection system. The smart patrol inspection system includes: a portable device, including: a camera module to obtain an environmental image; a wireless communication module to send associated One of environmental information in the environmental image and receiving a notification signal; a processor electrically connected to the camera module and the wireless communication module, and the processor is used to control the wireless communication module to send the environmental information; and a notification The output module is electrically connected to the processor, the notification output module is used to output the notification signal; and an artificial intelligence inspection and analysis platform, which is communicatively connected to the wireless communication module to receive the environmental information, the artificial intelligence inspection The analysis platform is used to load a three-dimensional model database and a route information, the route information includes a sequence of multiple coordinate points in the three-dimensional model database, the artificial intelligence inspection and analysis platform is based on the environmental information and the three-dimensional model data The library executes a comparison process to obtain a location information, and generates and sends the notification signal based on the location information and the route information. 如請求項1所述智慧巡檢系統,適用於一高速網路環境,該高速網路環境包括多個無線存取點,該些無線存取點應用第五代行動通訊技術(5th Generation mobile networks,5G)專用頻譜之企業專網彼此互相連接,且該無線通訊模組係透過該些無線存取點其中一者通訊連接至該人工智慧巡檢分析平台。The requested item wisdom of the inspection system 1 for a high-speed network environment, the network environment includes a plurality of high-speed wireless access point, wireless access point of the plurality of fifth-generation mobile communication technology applications (5 th Generation mobile Networks, 5G) dedicated spectrum enterprise private networks are connected to each other, and the wireless communication module is connected to the artificial intelligence inspection and analysis platform through one of the wireless access points. 如請求項1所述的智慧巡檢系統,其中該通知訊號係包含一路線影像,該通知輸出模組係一顯示裝置,該顯示裝置用以呈現該路線影像。The smart inspection system according to claim 1, wherein the notification signal includes a route image, the notification output module is a display device, and the display device is used for presenting the route image. 如請求項1所述的智慧巡檢系統,其中該通知訊號係包含一語音,該通知輸出模組係一揚聲器,該揚聲器用以播放該語音。The smart inspection system according to claim 1, wherein the notification signal includes a voice, and the notification output module is a speaker, and the speaker is used for playing the voice. 如請求項1所述的智慧巡檢系統,更包括:一慣性測量模組,電性連接該處理器,該慣性測量模組用以取得該可攜式裝置之一移動資訊;其中該環境影像係一第一環境影像,該環境資訊係一第一環境資訊,該位置資訊係一第一位置資訊;該處理器更用以控制該攝像模組在取得該第一環境影像並經過一間隔時間之後取得一第二環境影像;該無線通訊模組更用以發送關聯於該第二環境影像的一第二環境資訊及該移動資訊;該人工智慧巡檢分析平台更用以接收該第二環境資訊,依據該第二環境資訊、該三維模型資料庫、該間隔時間及該移動資訊執行該比對程序以獲得一第二位置資訊;並依據該第一位置資訊、該第二位置資訊及該路線資訊選擇性地產生一警示訊號。The smart inspection system according to claim 1, further comprising: an inertial measurement module electrically connected to the processor, and the inertial measurement module is used to obtain mobile information of the portable device; wherein the environmental image Is a first environment image, the environment information is a first environment information, and the location information is a first location information; the processor is further used to control the camera module to obtain the first environment image and pass an interval time Then obtain a second environment image; the wireless communication module is further used to send a second environment information and the mobile information related to the second environment image; the artificial intelligence inspection and analysis platform is further used to receive the second environment Information, according to the second environment information, the three-dimensional model database, the interval time and the movement information to execute the comparison procedure to obtain a second position information; and according to the first position information, the second position information and the The route information selectively generates a warning signal. 如請求項1所述的智慧巡檢系統,其中該三維模型資料庫包括一電腦輔助設計檔案。The smart inspection system according to claim 1, wherein the three-dimensional model database includes a computer-aided design file. 一種智慧巡檢方法,包括:以一可攜式裝置之一攝像模組取得一環境影像;以該可攜式裝置之一處理器控制該可攜式裝置之一無線通訊模組發送關聯於該環境影像之一環境資訊至一人工智慧巡檢分析平台;以該人工智慧巡檢分析平台載入一三維模型資料庫;以該人工智慧巡檢分析平台依據該環境資訊及該三維模型資料庫執行一比對程序以獲取一位置資訊;以該人工智慧巡檢分析平台依據該位置資訊及一路線資訊產生並發送該通知訊號至該可攜式裝置,其中該路線資訊包括該三維模型資料庫的多個座標點的一順序;以該可攜式裝置之該無線通訊模組接收該通知訊號;以及以該可攜式裝置之一通知輸出模組輸出該通知訊號。A smart inspection method includes: obtaining an environmental image by a camera module of a portable device; using a processor of the portable device to control a wireless communication module of the portable device to send an associated One of the environmental images of environmental information is sent to an artificial intelligence inspection and analysis platform; the artificial intelligence inspection and analysis platform is used to load a 3D model database; the artificial intelligence inspection and analysis platform is used to execute based on the environmental information and the 3D model database A comparison procedure is used to obtain a location information; the artificial intelligence inspection and analysis platform generates and sends the notification signal to the portable device according to the location information and a route information, wherein the route information includes the data of the 3D model database A sequence of a plurality of coordinate points; the wireless communication module of the portable device receives the notification signal; and a notification output module of the portable device outputs the notification signal. 如請求項7所述的智慧巡檢方法,其中該通知訊號係包含一路線影像,該通知輸出模組係一顯示裝置,且以該可攜式裝置之該通知輸出模組輸出該通知訊號係以該顯示裝置呈現該路線影像。The smart inspection method according to claim 7, wherein the notification signal includes a route image, the notification output module is a display device, and the notification output module of the portable device is used to output the notification signal. The route image is presented by the display device. 如請求項7所述的智慧巡檢方法,其中在以該可攜式裝置之該攝像模組取得該環境影像之前,更包括:以該可攜式裝置之該攝像模組取得一起點影像;在取得該起點影像之後,以該處理器控制該無線通訊模組發送該起點影像及一復歸訊號;以及以該人工智慧巡檢分析平台依據該復歸訊號及該起點影像執行該比對程序以獲得一確認訊號,並發送該確認訊號至該可攜式裝置。The smart inspection method according to claim 7, wherein before obtaining the environmental image with the camera module of the portable device, it further comprises: obtaining a point image with the camera module of the portable device; After obtaining the starting point image, use the processor to control the wireless communication module to send the starting point image and a reset signal; and use the artificial intelligence patrol analysis platform to execute the comparison procedure according to the reset signal and the starting point image to obtain A confirmation signal, and sending the confirmation signal to the portable device. 如請求項7所述的智慧巡檢方法,其中該處理器更用以儲存複數個預設影像,每一該些預設影像對應該些座標點中之一預設座標點,在以該可攜式裝置之該攝像模組取得該環境影像之前,更包括:以該攝像模組取得一起點影像;以該處理器比對該起點影像及每一該些預設影像;以及當該起點影像與該些預設影像之中一者相符,以該處理器控制該無線通訊模組發送對應於該些預設影像之中該者的該預設座標點。The smart inspection method according to claim 7, wherein the processor is further used to store a plurality of preset images, each of the preset images corresponds to one of the preset coordinate points, and the Before the camera module of the portable device obtains the environmental image, it further includes: obtaining a point image with the camera module; comparing the starting point image and each of the preset images with the processor; and when the starting point image In accordance with one of the preset images, the processor controls the wireless communication module to send the preset coordinate point corresponding to the one of the preset images.
TW109102386A 2020-01-22 2020-01-22 Artificial intelligence patrol system and artificial intelligence patrol method TWI730605B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW109102386A TWI730605B (en) 2020-01-22 2020-01-22 Artificial intelligence patrol system and artificial intelligence patrol method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW109102386A TWI730605B (en) 2020-01-22 2020-01-22 Artificial intelligence patrol system and artificial intelligence patrol method

Publications (2)

Publication Number Publication Date
TWI730605B TWI730605B (en) 2021-06-11
TW202129229A true TW202129229A (en) 2021-08-01

Family

ID=77517543

Family Applications (1)

Application Number Title Priority Date Filing Date
TW109102386A TWI730605B (en) 2020-01-22 2020-01-22 Artificial intelligence patrol system and artificial intelligence patrol method

Country Status (1)

Country Link
TW (1) TWI730605B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113916244A (en) * 2021-10-08 2022-01-11 江苏眸视机器人科技有限公司 Method and device for setting inspection position, electronic equipment and readable storage medium

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102915575A (en) * 2012-10-18 2013-02-06 中国工商银行股份有限公司 Electronic patrol monitoring device and system
TW201525945A (en) * 2013-12-24 2015-07-01 Hon Hai Prec Ind Co Ltd Patrol control device, system and method
TWM503610U (en) * 2014-12-24 2015-06-21 Sky Mirror Corp Patrol device
CN106097471A (en) * 2016-06-02 2016-11-09 深圳朝伟达科技有限公司 A kind of instant security personnel's patrol system and method
GR1009254B (en) * 2017-03-17 2018-03-23 Egt Μονοπροσωπη Εταιρεια Περιορισμενης Ευθυνης Με Δ.Τ. "Egt Ltd" Methodology and integrated guard tour patrol system
TWM594124U (en) * 2020-01-22 2020-04-21 國眾電腦股份有限公司 Artificial intelligence patrol system

Also Published As

Publication number Publication date
TWI730605B (en) 2021-06-11

Similar Documents

Publication Publication Date Title
US20210019627A1 (en) Target tracking method and apparatus, medium, and device
CN106595565B (en) A kind of intellectual faculties based on BIM
KR102289745B1 (en) System and method for real-time monitoring field work
US11111785B2 (en) Method and device for acquiring three-dimensional coordinates of ore based on mining process
EP2973432A1 (en) Method and system for disambiguation of augmented reality tracking databases
CN112115607B (en) Mobile intelligent body digital twin system based on multidimensional microblog space
US20190385364A1 (en) Method and system for associating relevant information with a point of interest on a virtual representation of a physical object created using digital input data
CN114584571B (en) Space calculation technology-based digital twin synchronous communication method for power grid station
CN111988524A (en) Unmanned aerial vehicle and camera collaborative obstacle avoidance method, server and storage medium
JP2024020295A (en) Cooperative virtual interface
EP4336385A1 (en) Method and apparatus for updating target detection model
CN112562092A (en) Electrical test intelligent interaction system based on AR technology
US20140180743A1 (en) Workflow monitoring and analysis system and method thereof
US11395102B2 (en) Field cooperation system and management device
Feng et al. Three-dimensional robot localization using cameras in wireless multimedia sensor networks
TWI730605B (en) Artificial intelligence patrol system and artificial intelligence patrol method
TWI773898B (en) Management apparatus and management method thereof for electronic equipment
CN112667139A (en) Robot operation method, device, equipment and storage medium based on augmented reality
TWM594124U (en) Artificial intelligence patrol system
Oh et al. Monocular UAV localisation with deep learning and uncertainty propagation
Agrawal et al. RescueAR: Augmented reality supported collaboration for UAV driven emergency response systems
JP6725736B1 (en) Image specifying system and image specifying method
EP4216100A1 (en) Architecture for distributed artificial intelligence augmentation
CN115797406A (en) Out-of-range warning method, device, equipment and storage medium
US20220164492A1 (en) Methods and apparatus for two dimensional location based digital content