TWM600873U - Detection system for detecting road damage - Google Patents

Detection system for detecting road damage Download PDF

Info

Publication number
TWM600873U
TWM600873U TW109203029U TW109203029U TWM600873U TW M600873 U TWM600873 U TW M600873U TW 109203029 U TW109203029 U TW 109203029U TW 109203029 U TW109203029 U TW 109203029U TW M600873 U TWM600873 U TW M600873U
Authority
TW
Taiwan
Prior art keywords
gnss
rtk
detection system
road
image
Prior art date
Application number
TW109203029U
Other languages
Chinese (zh)
Inventor
游勳喬
吳順德
章皓鈞
Original Assignee
日陞空間資訊股份有限公司
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 日陞空間資訊股份有限公司 filed Critical 日陞空間資訊股份有限公司
Priority to TW109203029U priority Critical patent/TWM600873U/en
Publication of TWM600873U publication Critical patent/TWM600873U/en

Links

Images

Landscapes

  • Image Analysis (AREA)
  • Traffic Control Systems (AREA)

Abstract

本創作提供一種用於檢測道路破損之檢測系統，其包括：一GNSS-RTK 定位設備，用於接收與傳輸觀測資料；一移動載具，其包含：一影像擷取設備，用於擷取道路鋪面之影像；一GNSS-RTK移動設備，用以接收與傳輸該GNSS-RTK定位設備之觀測資料；一計算單元，資訊連接於該影像擷取設備與該GNSS-RTK移動設備，根據該影像辨別道路破損，並同時根據該觀測資料計算與儲存該GNSS-RTK移動設備瞬時座標，來達到精準定位來檢測及記錄道路破損。 This creation provides a detection system for detecting road damage, which includes: a GNSS-RTK Positioning equipment for receiving and transmitting observation data; a mobile vehicle, which includes: an image capturing device for capturing images of road paving; a GNSS-RTK mobile device for receiving and transmitting the GNSS-RTK Observation data of positioning equipment; a computing unit, which is connected to the image capturing equipment and the GNSS-RTK mobile device, identifies road damage based on the image, and calculates and stores the instantaneous coordinates of the GNSS-RTK mobile device based on the observation data , To achieve precise positioning to detect and record road damage.

Description

用於檢測道路破損之檢測系統 Detection system for detecting road damage

本創作係關於一種檢測道路破損之檢測系統，尤其是利用多衛星組合結合實時動態測量技術(Global Navigation Satellite System-Real Time Kinematic，GNSS-RTK)以及擷取影像方式，來檢測道路破損、紀錄破損座標位置以及預估破損之大小之系統。 This creation is about a road damage detection system, especially the use of multi-satellite combination combined with real-time dynamic measurement technology (Global Navigation Satellite System-Real Time Kinematic, GNSS-RTK) and image capture methods to detect road damage and record damage Coordinate position and system for estimating the size of damage.

道路是現今社會的物流命脈，也是與人民日常通勤息息相關的，因此路面的平整與否不僅會對經濟有所影響，也對人民的交通安全有重大的關係。但是道路使用久了，破損的產生是無可避免的，而目前國內道路維護多為定期派遣工程車檢測與人民通報等方式，但此方式甚為花費人力與時間，此外每個人對於道路破損判斷標準皆有不同，因此希望借助科技以達到客觀的判斷。 Roads are the lifeblood of logistics in today's society and are closely related to people's daily commuting. Therefore, the smoothness of the roads will not only affect the economy, but also have a major relationship with the people's traffic safety. However, the road has been used for a long time, and damage is inevitable. At present, domestic road maintenance is mostly carried out by regularly dispatching engineering vehicles for inspection and people’s notification, but this method is very labor and time-consuming. In addition, everyone judges the road damage The standards are different, so we hope to use technology to achieve objective judgments.

隨著科技的發展，AI領域逐漸地成熟，深度學習已經運用在許多領域上，像是：癌症診斷、設備診斷、金融營銷等，但是目前還未使用於路面檢測，透過深度學習取得影像中物件的細微特徵，為分析路面破損有更加精確的方式。 With the development of science and technology, the field of AI has gradually matured. Deep learning has been used in many fields, such as cancer diagnosis, equipment diagnosis, financial marketing, etc., but it has not yet been used in road detection. Objects in images are obtained through deep learning. The subtle features of, provide a more accurate way to analyze road damage.

目前巡查車輛係於車上裝設有GPS定位裝置及攝影機，巡查人員可將巡查路線全程進行攝影，巡查過程中若發現道路破損，則可利用車上電腦即時將破損進行標定截取畫面，並利用GPS定位確認破損座標，透過無線傳輸將 查報資訊傳至監控中心，由於使用人工判斷，因此有可能產生漏判、判斷錯誤的可能，若使用AI進行自動檢測，則可避免此問題，並且減少人力的花費；此外，因為GPS有受以下問題影響而導致在座標定位上會有較大的誤差。 At present, the inspection vehicle is equipped with a GPS positioning device and a camera. The inspector can take pictures of the entire inspection route. If the road is damaged during the inspection, the on-board computer can be used to calibrate the damage in real time to capture the screen and use GPS positioning to confirm the damaged coordinates, through wireless transmission The report information is transmitted to the monitoring center. Due to the use of manual judgment, there may be missed judgments and judgment errors. If AI is used for automatic detection, this problem can be avoided and labor costs can be reduced; in addition, because GPS is affected The following problems will cause large errors in coordinate positioning.

一、大氣層影響：大氣層中的電離層和對流層對電磁波的折射效應，使得GPS信號的傳播速度發生變化，從而讓GPS信號產生延遲。 1. Atmospheric influence: The refraction effect of the ionosphere and troposphere on electromagnetic waves in the atmosphere changes the propagation speed of GPS signals, thereby delaying GPS signals.

二、衛星星曆誤差：由於衛星運行中受到複雜的外力作用，而地面控制站和接收終端無法測定和掌握其規律，從而無法消除產生的誤差。 2. Satellite ephemeris error: Due to the complicated external force during the operation of the satellite, the ground control station and receiving terminal cannot determine and master the law, so the error cannot be eliminated.

三、多路徑效應：GPS信號在不同的障礙物上反射後才被接收到，而使時間產生延遲。 3. Multipath effect: GPS signals are received after being reflected on different obstacles, which causes time delay.

由於上述因素，使GPS技術的應用在某種程度上受到了限制，為了達到一定的精度，在實際應用中往往會要求與地面控制站的距離不能太長(需小於10公里)。 Due to the above factors, the application of GPS technology is limited to some extent. In order to achieve a certain accuracy, the distance from the ground control station is often required to not be too long (less than 10 kilometers) in actual applications.

此外，行駛於高架橋下方、隧道、地下停車場等有遮掩物的環境時，因無法即時獲得定位資訊，而造成無法定位等問題。 In addition, when driving under a viaduct, tunnels, underground parking lots and other sheltered environments, the positioning information cannot be obtained in real time, causing problems such as inability to locate.

為解決上述之定位精準度不良之缺失以及辨別路面破損，本創作提供一種用於檢測道路破損之檢測系統，其包括：一GNSS-RTK定位設備，用於接收與傳輸觀測資料；一移動載具，其包含：一影像擷取設備，用於擷取道路鋪面之影像；一GNSS-RTK移動設備，用以接收與傳輸該GNSS-RTK定位設備之觀測資料；一計算單元，資訊連接於該影像擷取設備與該GNSS-RTK移動設備，根 據該影像辨別道路破損，並同時根據該觀測資料計算與儲存該GNSS-RTK移動設備瞬時座標。 In order to solve the above-mentioned lack of poor positioning accuracy and identify road damage, this creation provides a detection system for detecting road damage, which includes: a GNSS-RTK positioning device for receiving and transmitting observation data; a mobile vehicle , Which includes: an image capturing device for capturing images of road paving; a GNSS-RTK mobile device for receiving and transmitting observation data of the GNSS-RTK positioning device; a computing unit with information connected to the image The capture device and the GNSS-RTK mobile device, root Identify road damage based on the image, and calculate and store the instantaneous coordinates of the GNSS-RTK mobile device based on the observation data.

如上述之用於檢測道路破損之檢測系統，其中該移動載具更包括一慣性導航模組，資訊連接於該計算單元。 As in the above-mentioned detection system for detecting road damage, the mobile vehicle further includes an inertial navigation module, and the information is connected to the calculation unit.

如上述之用於檢測道路破損之檢測系統，其中，該影像擷取設備包含至少一個攝影機。 As in the above-mentioned detection system for detecting road damage, the image capturing device includes at least one camera.

如上述之用於檢測道路破損之檢測系統，其中該計算單元使用深度學習訓練模型來辨識影像中的路面是否有破損。 As in the above-mentioned detection system for detecting road damage, the calculation unit uses a deep learning training model to identify whether the road surface in the image is damaged.

如上述之用於檢測道路破損之檢測系統，其中該深度學習訓練模型可包含區域卷積神經網路(Region-based Convolutional Neural Network，R-CNN)、快速區域卷積神經網路(Fast Region-based Convolutional Neural Network，Fast R-CNN)、更快速區域卷積神經網路(Faster Region-based Convolutional Neural Network，Faster R-CNN)、遮罩區域卷積神經網路模型之模型(Mask Region-based Convolutional Neural Network，Mask R-CNN)其中之一者。 Such as the above-mentioned detection system for detecting road damage, wherein the deep learning training model may include a Region-based Convolutional Neural Network (R-CNN) and a Fast Region-based Convolutional Neural Network (R-CNN). based Convolutional Neural Network, Fast R-CNN), Faster Region-based Convolutional Neural Network (Faster R-CNN), Mask Region-based Convolutional Neural Network model (Mask Region-based) Convolutional Neural Network, Mask R-CNN) one of them.

如上述之用於檢測道路破損之檢測系統，其中該深度學習訓練模型可為YOLO(You Only Look Once)的檢測框架或單次多框偵測器(Single Shot Multibox Detector，SSD)的檢測框架。 As in the detection system for detecting road damage, the deep learning training model can be the detection framework of YOLO (You Only Look Once) or the detection framework of Single Shot Multibox Detector (SSD).

如上述之用於檢測道路破損之檢測系統，其中該計算單元資訊連接於雲端或者網際網路，由該雲端替代該計算單元，根據該影像辨別道路破損，並同時根據該觀測資料計算與儲存該GNSS-RTK移動設備瞬時座標。 Such as the above-mentioned detection system for detecting road damage, wherein the computing unit information is connected to the cloud or the Internet, and the computing unit is replaced by the cloud, the road damage is identified based on the image, and the observation data is calculated and stored at the same time GNSS-RTK mobile device instantaneous coordinates.

100:路面破損檢測之系統 100: Pavement damage detection system

101:移動載具 101: Mobile Vehicle

102:影像擷取裝置 102: Image capture device

103:GNSS-RTK移動設備 103: GNSS-RTK mobile device

104:計算單元 104: calculation unit

105:慣性導航模組 105: Inertial Navigation Module

200:GNSS-RTK定位設備 200: GNSS-RTK positioning equipment

300:衛星 300: Satellite

D:焦距至路面破損的距離 D: Distance from focal length to road damage

FL:焦距 FL: Focal length

IM:道路影像/路面破損大小 IM: Road image/size of road damage

LC:左鏡頭 LC: Left lens

RC:右鏡頭 RC: Right lens

S:感光元件大小 S: photosensitive element size

圖1顯示為本創作路面破損檢測系統的示意圖。 Figure 1 shows a schematic diagram of this creative road damage detection system.

圖2顯示為本創作GNSS-RTK技術之示意圖。 Figure 2 shows a schematic diagram of this creative GNSS-RTK technology.

圖3顯示為本創作移動載具檢測時的示意圖。 Figure 3 shows a schematic diagram of the mobile vehicle detection for this creation.

圖4顯示為本創作計算單元之工作流程圖。 Figure 4 shows the work flow chart of this authoring calculation unit.

圖5顯示為本創作Mask R-CNN模型之流程圖。 Figure 5 shows the flow chart for creating the Mask R-CNN model.

圖6顯示為本創作Mask R-CNN訓練模型之流程圖。 Figure 6 shows the flow chart of the Mask R-CNN training model for this creation.

圖7顯示為本創作Mask R-CNN預測模型之流程圖。 Figure 7 shows the flow chart of this creation of the Mask R-CNN prediction model.

圖8顯示為本創作GNSS-RTK技術之流程圖。 Figure 8 shows the flow chart of the GNSS-RTK technology for this creation.

圖9顯示為本創作破損大小計算方法之示意圖。 Figure 9 shows a schematic diagram of the method for calculating the damage size of this creation.

圖10顯示為本創作雙鏡頭測距之示意圖。 Figure 10 shows a schematic diagram of this creative dual-lens distance measurement.

本創作之實施方式將於下文中，參照本創作的理想實施方式的示意圖來進行描述。該等圖示中的形狀、設置方式會因製造技術、設計及/或公差而有所不同。因此，本創作文中所說明的實施方式不應被視為是用來將本創作結構侷限在特定的元件或形狀，其應包含任何因製作所造成在形狀方面的差異。 The implementation of this creation will be described below with reference to the schematic diagram of the ideal implementation of this creation. The shapes and setting methods in these diagrams may vary due to manufacturing technology, design and/or tolerances. Therefore, the implementation described in this creation article should not be regarded as limiting the creation structure to specific elements or shapes, and should include any difference in shape caused by production.

請參閱圖1之路面破損檢測之系統100的示意圖，如圖所示，當移動載具101在道路上進行檢測道路是否有破損，其較為深色的區域為路面破損檢測的範圍，其中該檢測範圍為不限定，因此360度檢測道路也是可行的。 Please refer to the schematic diagram of the road damage detection system 100 in Figure 1. As shown in the figure, when the mobile vehicle 101 is on the road to detect whether the road is damaged, the darker area is the range of the road damage detection. The range is not limited, so 360-degree road detection is also feasible.

請參閱本創作圖2之GNSS-RTK技術之示意圖，如圖所示，GNSS-RTK技術包含：複數個衛星300，以及地面上設有複數個定點的GNSS-RTK定位設備200(為方便敘述，只用單個圖示)與設置於移動載具101上的GNSS-RTK移動設備103，其中該GNSS-RTK定位設備200最佳設置為架設在已知高精度座標的點位上，也就是作為能接收複數個衛星300所傳送的觀測資料，以及透過無線電設備的傳輸，將該GNSS-RTK定位設備200的觀測資料傳送給該GNSS-RTK移動設備103，因此該GNSS-RTK移動設備103不限定設置於該移動載具101上方，而能接收該衛星300以及該GNSS-RTK定位設備200的觀測資料之設置位置即可，為方便理解下文之內容，此處只說明GNSS-RTK技術的連接關係，其關於GNSS-RTK技術的詳細流程將後述。 Please refer to the schematic diagram of the GNSS-RTK technology in Figure 2 of this creation. As shown in the figure, the GNSS-RTK technology includes: a plurality of satellites 300, and a GNSS-RTK positioning device 200 with a plurality of fixed points on the ground (for ease of description, Only a single illustration) and the GNSS-RTK mobile device 103 set on the mobile vehicle 101, where the GNSS-RTK positioning device 200 is best set to be erected on a point with known high-precision coordinates, that is, as an energy source Receive observation data transmitted by a plurality of satellites 300, and transmit the observation data of the GNSS-RTK positioning device 200 to the GNSS-RTK mobile device 103 through the transmission of radio equipment, so the GNSS-RTK mobile device 103 is not limited to settings Just above the mobile vehicle 101, it is enough to receive the observation data of the satellite 300 and the GNSS-RTK positioning device 200. To facilitate the understanding of the content below, only the connection relationship of the GNSS-RTK technology is explained here. The detailed process of GNSS-RTK technology will be described later.

接著，請參閱本創作圖3之移動載具檢測時的示意圖，該行動載具101搭載影像擷取設備102，如：攝影機、GNSS-RTK移動設備103與計算單元104，其中該影像擷取設備102、該GNSS-RTK移動設備103與該計算單元104可/或為一體設置，該計算單元104資訊連接於該影像擷取設備102以及該GNSS-RTK移動設備103，首先，當該行動載具101於路面上遇到破損的路面時，由該影像擷取裝置102將檢測範圍內所接收之道路影像IM進行擷取，以及根據該GNSS-RTK移動設備103從該GNSS-RTK定位設備200所得到的觀測資料，接著同時將該道路影像IM以及該座標資料傳輸至該計算單元104進行道路影像IM辨別處理、紀錄當前座標、估測路面破損大小之處理程序。 Next, please refer to the schematic diagram of mobile vehicle detection in Figure 3 of this creation. The mobile vehicle 101 is equipped with an image capture device 102, such as a camera, a GNSS-RTK mobile device 103, and a computing unit 104, wherein the image capture device 102. The GNSS-RTK mobile device 103 and the calculation unit 104 can be integrated, and the calculation unit 104 is connected to the image capturing device 102 and the GNSS-RTK mobile device 103. First, when the mobile vehicle When 101 encounters a damaged road on the road, the image capturing device 102 captures the road image IM received within the detection range, and according to the GNSS-RTK mobile device 103 from the GNSS-RTK positioning device 200 The obtained observation data is then simultaneously transmitted to the calculation unit 104 to perform road image IM identification processing, recording current coordinates, and estimating the size of road damage.

接著，將對於路面破損檢測之系統100的步驟與方法的細節進行說明，首先，請參閱圖4計算單元104之工作流程圖，如圖所示，該計算單元104之 處理流程共有5個步驟，其中根據不同的需求，流程圖中所示步驟的執行順序可以調整或者部分步驟可以省略。 Next, the details of the steps and methods of the road damage detection system 100 will be described. First, please refer to the flowchart of the calculation unit 104 in FIG. 4, as shown in the figure, the calculation unit 104 There are 5 steps in the processing flow. According to different requirements, the order of execution of the steps shown in the flowchart can be adjusted or some steps can be omitted.

步驟S101，該計算單元104從該影像擷取裝置102讀入擷取之路面影像IM。 In step S101, the calculation unit 104 reads the captured road image IM from the image capturing device 102.

步驟S102，根據檢測路面影像IM，該計算單元104利用深度學習模型進行快速且統一的辨別路面是否有破損，以下對於破損辨別的訓練與預測的模型進行說明，本創作最佳的深度學習模型配置為使用遮罩區域卷積神經網路模型(Mask Region based Convolutional Neural Network，Mask R-CNN)，當然也可使用習知的深度學習模型，如：區域卷積神經網路(Region based Convolutional Neural Network，R-CNN)、快速區域卷積神經網路(Fast Region based Convolutional Neural Network，Fast R-CNN)、更快速區域卷積神經網路(Faster Region based Convolutional Neural Network，Faster R-CNN)，或者使用YOLO(You Only Look Once)的檢測框架或單次多框偵測器(Single Shot Multibox Detector，SSD)的檢測框架進行影像辨識，請同時參閱圖5至圖7，圖5為本創作Mask R-CNN模型之流程圖，圖6為本創作Mask R-CNN訓練模型之流程圖，圖7為本創作Mask R-CNN檢測模型之流程圖。本實施例使用骨幹(Backbone)為殘差網路101(Residual Network 101，ResNet101)之Mask R-CNN訓練模型之流程圖進行檢測路面破損。 Step S102, according to the detected road image IM, the calculation unit 104 uses the deep learning model to quickly and uniformly identify whether the road is damaged. The following describes the damage identification training and prediction model. The best deep learning model configuration for this creation In order to use the Mask Region based Convolutional Neural Network (Mask R-CNN), of course, the well-known deep learning model can also be used, such as: Region based Convolutional Neural Network (Region based Convolutional Neural Network) , R-CNN), Fast Region based Convolutional Neural Network (Fast R-CNN), Faster Region based Convolutional Neural Network (Faster R-CNN), or Use the detection framework of YOLO (You Only Look Once) or the detection framework of Single Shot Multibox Detector (SSD) for image recognition. Please also refer to Figures 5 to 7. Figure 5 is the creation of Mask R -The flow chart of the CNN model. Figure 6 is the flow chart of creating the Mask R-CNN training model. Figure 7 is the flow chart of creating the Mask R-CNN detection model. In this embodiment, the flowchart of the Mask R-CNN training model in which the backbone is the residual network 101 (Residual Network 101, ResNet101) is used to detect road damage.

本創作Mask R-CNN模型之流程圖為一種物件檢測及實例分割(Instance Segmentation)模型，步驟S201，將影像輸入到一個預訓練好的卷積神經網絡中進行特徵提取，獲得對應的特徵圖(feature map)，步驟S202，使用區域候選網路(Region Proposal Network，RPN)在feature map提取出候選框(region proposals)，並以候選框分數篩選出準確度較高的感興趣區域(Region of interest，RoI)，步驟S203，使用RoI Align層提取這些RoI的特徵，步驟S204，對每個採樣的RoI定義一個多任務損失函數Loss Function=類別損失(Classification Loss)+邊框回歸損失(Bounding box regression Loss)+遮罩損失(Mask Loss)，其中訓練時，更快速區域卷積神經網路模型(Faster Region based Convolutional Neural Network，Faster R-CNN)分支與遮罩分支是分開並行訓練。使用Mask R-CNN模型進行檢測步驟與訓練相似，請參閱圖6與7，步驟S201至步驟S203與圖5皆相同，而不同於圖5的步驟S204，訓練時Faster R-CNN分支與遮罩分支是分開並行訓練，而預測時，先運行Faster R-CNN分支獲得步驟S203之RoI具體Classification(如：龜裂、坑洞等)與Bounding box後，再運行遮罩分支，來預測每個感興趣區域(RoI)上的分割蒙版，即可快速的檢測出破損類別，同時自動圈選出破損區域，並將破損圖片與破損訊息儲存於該計算單元104中。 The flowchart of this creation of Mask R-CNN model is an object detection and instance segmentation (Instance Segmentation) model. In step S201, the image is input into a pre-trained convolutional neural network for feature extraction to obtain the corresponding feature map ( feature map), step S202, using a region candidate network (Region Proposal Network, RPN) to extract a candidate frame (region proposals), and use the candidate frame scores to filter out regions of interest (Region of interest, RoI) with higher accuracy. Step S203, use the RoI Align layer to extract the features of these RoIs, and step S204, define one for each sampled RoI Multi-task loss function Loss Function=Classification Loss+Bounding box regression Loss+Mask Loss, in which, during training, the faster region based convolutional neural network model (Faster Region based Convolutional Neural Network, Faster R-CNN) branch and mask branch are separately trained in parallel. Using the Mask R-CNN model to perform detection steps is similar to training. Please refer to Figures 6 and 7. Steps S201 to S203 are the same as Figure 5, but different from step S204 in Figure 5. Faster R-CNN branch and mask during training The branches are trained separately and in parallel. When predicting, first run the Faster R-CNN branch to obtain the specific RoI Classification (such as cracks, potholes, etc.) and Bounding box in step S203, and then run the mask branch to predict each sense The segmentation mask on the region of interest (RoI) can quickly detect the damage category, and at the same time automatically circle the damaged area, and store the damaged image and damage information in the calculation unit 104.

請參閱圖4之計算單元104之工作流程圖步驟S103，運用GNSS-RTK技術獲取高精度的座標位置並記錄，請參閱圖8，並同時參閱圖2所示，其方法為：利用該GNSS-RTK定位設備200將已知該GNSS-RTK定位設備200的座標與載波相位觀測量等資料(接收觀測時瞬間自行產生的定位相位與該衛星300接收到的相位差)，透過通訊設備將該觀測資料即時傳送給該GNSS-RTK移動設備103，該GNSS-RTK移動設備103再經由OTF(On-the-Fly)週波未定值搜尋法快速解算週波未定值(該衛星300與該GNSS-RTK定位設備200之間的整數週波值為未知值，稱為週波未定值，其中，OTF泛指在移動的狀態下，正確求解週波未定值之演算法)，之後利用差分定位，再計算GNSS-RTK移動設備103的瞬時座標， 意即GNSS-RTK移動設備103可以在純動態環境下求解週波未定值，另外，兩站之間距離越近，越能消除該GNSS-RTK定位設備200與該GNSS-RTK移動設備103間的共同性誤差，而週波未定值解算時間也越短，定位精度越高。 Please refer to step S103 of the work flow chart of the calculation unit 104 in FIG. 4, use GNSS-RTK technology to obtain high-precision coordinate positions and record them. Please refer to FIG. 8 and also refer to FIG. 2. The method is: use the GNSS- The RTK positioning device 200 will know the coordinates and carrier phase observation data of the GNSS-RTK positioning device 200 (the positioning phase generated by itself at the moment of receiving the observation and the phase difference received by the satellite 300), and the observation will be made through the communication device The data is sent to the GNSS-RTK mobile device 103 in real time, and the GNSS-RTK mobile device 103 uses the OTF (On-the-Fly) cycle undetermined value search method to quickly calculate the undetermined value (the satellite 300 and the GNSS-RTK positioning The integer cycle value between the devices 200 is an unknown value, which is called cycle undetermined value. Among them, OTF generally refers to the algorithm that correctly solves the cycle undetermined value in a moving state, and then uses differential positioning to calculate the GNSS-RTK movement The instantaneous coordinates of device 103, This means that the GNSS-RTK mobile device 103 can solve the undetermined frequency in a purely dynamic environment. In addition, the closer the distance between the two stations, the more the commonality between the GNSS-RTK positioning device 200 and the GNSS-RTK mobile device 103 can be eliminated. The shorter the solution time of the undetermined value of the cycle, the higher the positioning accuracy.

故使用GNSS-RTK技術，可以去除掉該GNSS-RTK移動設備103與該GNSS-RTK定位設備200間的共同誤差以及獲得該GNSS-RTK移動站之釐米級瞬時坐標，相對於GPS，其定位精度相對甚大，因此對於道路破損的地方更能精準定位，而不會發生施工人員在道路破損的座標上找不到破損路面的情況。 Therefore, using the GNSS-RTK technology, the common error between the GNSS-RTK mobile device 103 and the GNSS-RTK positioning device 200 can be removed and the centimeter-level instantaneous coordinates of the GNSS-RTK mobile station can be obtained. Relative to GPS, its positioning accuracy It is relatively large, so it can more accurately locate the damaged road, and it will not happen that the construction personnel can not find the damaged road surface on the coordinate of the road damage.

此外，本創作檢測道路破損之系統另具備慣性導航模組105(未圖示)，當行駛於高架橋下方、隧道、地下停車場等有遮掩物的環境時，GNSS定位精度大大的降低或是無法定位，使用該慣性導航模組105，即使在GNSS信號丟失的駕駛過程中也可以準確的定位，慣性導航技術使用陀螺儀用來形成一個導航坐標系，使加速度計的測量軸穩定在該坐標系中，並給出航向和姿態角；加速度計用來測量運動體的加速度，經過對時間的一次積分得到速度，速度再經過對時間的一次積分即可得到位移，如此一來，當GNSS信號丟失時，可以預測出即時的座標位置。 In addition, the system for detecting road damage in this creation is equipped with an inertial navigation module 105 (not shown). When driving under an overpass, tunnel, underground parking lot, etc., the GNSS positioning accuracy is greatly reduced or cannot be located. , Using the inertial navigation module 105, it can be accurately positioned even during driving when the GNSS signal is lost. Inertial navigation technology uses a gyroscope to form a navigation coordinate system to stabilize the measurement axis of the accelerometer in the coordinate system , And give the heading and attitude angle; the accelerometer is used to measure the acceleration of the moving body, and the speed is obtained after one integration of time, and the displacement can be obtained after one time integration of the speed. In this way, when the GNSS signal is lost , Can predict the real-time coordinate position.

步驟S104，運用攝影測量預估破損大小IM，如圖9所示，藉由該影像擷取設備102的焦距FL與感光元件S正比於焦點至破損的距離D與路面破損大小IM，其中該影像擷取設備102與影像中路面距離由雙鏡頭(對應於圖10的左鏡頭LC與右鏡頭RC)測距得到，接著請參閱圖10，根據三角形相似定律：

Figure 109203029-A0305-02-0009-1
Figure 109203029-A0305-02-0009-2
，由式(1)解得方程式：
Figure 109203029-A0305-02-0009-3
，
Figure 109203029-A0305-02-0009-5
，可得像機距P之距離為
Figure 109203029-A0305-02-0010-6
，其中f為相機焦距，P(x，z)為目標座標，此外焦距與感光元件長度為已知，因此可以計算得到破損之大小。 Step S104, use photogrammetry to estimate the damage size IM. As shown in FIG. 9, the focal length FL and the photosensitive element S of the image capturing device 102 are proportional to the distance D from the focus to the damage and the road damage size IM, where the image The distance between the capture device 102 and the road surface in the image is obtained by distance measurement with dual lenses (corresponding to the left lens LC and the right lens RC in FIG. 10). Then, referring to FIG. 10, according to the triangle similarity law:
Figure 109203029-A0305-02-0009-1
Figure 109203029-A0305-02-0009-2
, The equation is solved by equation (1):
Figure 109203029-A0305-02-0009-3
,
Figure 109203029-A0305-02-0009-5
, The available distance between the camera and P is
Figure 109203029-A0305-02-0010-6
, Where f is the focal length of the camera, P(x, z) is the target coordinate, and the focal length and the length of the photosensitive element are known, so the size of the damage can be calculated.

步驟S105，將破損圖片、破損座標與破損估算大小資訊彙整、儲存於該計算單元中104，以利道路破損管理，或者該計算單元中104資訊連接於雲端或者網際網路，該計算單元中104也可將該觀測資料、該道路影像IM上傳至雲端或者網際網路，由雲端或者網際網路代替該計算單元中104進行上述之深度學習對於影像辨別、訓練以及瞬時座標等計算儲存，更佳的是，可以做即時的管理，配合即時的管理系統，當附近有維修工程車，可以直接或者算出最佳的維修排定行程進行維修，因此可減少人力時間的付出，進而增加效率。 In step S105, the damaged image, damage coordinates, and damage estimated size information are collected and stored in the computing unit 104 to facilitate road damage management, or the information 104 in the computing unit is connected to the cloud or the Internet, 104 in the computing unit The observation data and the road image IM can also be uploaded to the cloud or the Internet, and the computing unit 104 can be replaced by the cloud or the Internet to perform the above-mentioned deep learning. It is better for image identification, training, and instantaneous coordinate calculation and storage. What's more, it can do real-time management and cooperate with real-time management system. When there is a maintenance engineering vehicle nearby, you can directly or calculate the best maintenance schedule for maintenance, so it can reduce the labor time and increase efficiency.

103:GNSS-RTK移動設備 103: GNSS-RTK mobile device

200:GNSS-RTK定位設備 200: GNSS-RTK positioning equipment

300:衛星 300: Satellite

Claims (8)

一種用於檢測道路破損之檢測系統，其包括：一GNSS-RTK(Global Navigation Satellite System-Real Time Kinematic)定位設備，用於接收與傳輸一觀測資料；一移動載具，其包含：一影像擷取設備，用於擷取一道路鋪面之影像；一GNSS-RTK移動設備，用以接收與傳輸該GNSS-RTK定位設備之觀測資料；一計算單元，資訊連接於該影像擷取設備與該GNSS-RTK移動設備，根據該影像辨別道路破損，並同時根據該觀測資料計算與儲存該GNSS-RTK移動設備瞬時座標。 A detection system for detecting road damage, including: a GNSS-RTK (Global Navigation Satellite System-Real Time Kinematic) positioning device for receiving and transmitting an observation data; a mobile vehicle, which includes: an image capture The acquisition device is used to capture an image of a road pavement; a GNSS-RTK mobile device is used to receive and transmit the observation data of the GNSS-RTK positioning device; a computing unit is used to connect the image capture device and the GNSS -RTK mobile device, distinguish road damage based on the image, and calculate and store the instantaneous coordinates of the GNSS-RTK mobile device based on the observation data. 如請求項1所述之檢測系統，其中該移動載具更包括一慣性導航模組，資訊連接於該計算單元。 The detection system according to claim 1, wherein the mobile vehicle further includes an inertial navigation module, and the information is connected to the calculation unit. 如請求項1所述之檢測系統，其中該觀測資料為一載波相位觀測資料。 The detection system according to claim 1, wherein the observation data is a carrier phase observation data. 如請求項1所述之檢測系統，其中，該影像擷取設備包含至少一個攝影機。 The detection system according to claim 1, wherein the image capturing device includes at least one camera. 如請求項1所述之檢測系統，其中該計算單元使用一深度學習訓練模型來辨識該道路鋪面之影像中的路面是否有破損。 The detection system according to claim 1, wherein the calculation unit uses a deep learning training model to identify whether the road surface in the image of the road pavement is damaged. 如請求項5所述之檢測系統，其中該深度學習訓練模型可包含區域卷積神經網路(Region based Convolutional Neural Network，R-CNN)、快 速區域卷積神經網路(Fast Region based Convolutional Neural Network，Fast R-CNN)、更快速區域卷積神經網路(Faster Region based Convolutional Neural Network，Faster R-CNN)、遮罩區域卷積神經網路模型之模型(Mask Region based Convolutional Neural Network，Mask R-CNN)其中之一者。 The detection system according to claim 5, wherein the deep learning training model may include a regional convolutional neural network (Region based Convolutional Neural Network, R-CNN), fast Fast Region based Convolutional Neural Network (Fast Region based Convolutional Neural Network, Fast R-CNN), Faster Region based Convolutional Neural Network (Faster R-CNN), Masked Region Convolutional Neural Network One of the model of the road model (Mask Region based Convolutional Neural Network, Mask R-CNN). 如請求項5所述之檢測系統，其中該深度學習訓練模型可為YOLO(You Only Look Once)的檢測框架或單次多框偵測器(Single Shot Multibox Detector，SSD)的檢測框架。 The detection system according to claim 5, wherein the deep learning training model can be a detection framework of YOLO (You Only Look Once) or a detection framework of Single Shot Multibox Detector (SSD). 如請求項1所述之檢測系統，其中該計算單元資訊連接於雲端或者網際網路，由該雲端替代該計算單元，根據該影像辨別道路破損，並同時根據該觀測資料計算與儲存該GNSS-RTK移動設備瞬時座標。 The detection system according to claim 1, wherein the computing unit information is connected to the cloud or the Internet, the computing unit is replaced by the cloud, the road damage is identified based on the image, and the GNSS- RTK mobile device instantaneous coordinates.
TW109203029U 2020-03-17 2020-03-17 Detection system for detecting road damage TWM600873U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW109203029U TWM600873U (en) 2020-03-17 2020-03-17 Detection system for detecting road damage

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW109203029U TWM600873U (en) 2020-03-17 2020-03-17 Detection system for detecting road damage

Publications (1)

Publication Number Publication Date
TWM600873U true TWM600873U (en) 2020-09-01

Family

ID=73644815

Family Applications (1)

Application Number Title Priority Date Filing Date
TW109203029U TWM600873U (en) 2020-03-17 2020-03-17 Detection system for detecting road damage

Country Status (1)

Country Link
TW (1) TWM600873U (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI765558B (en) * 2021-02-02 2022-05-21 覺華工程科技股份有限公司 Intelligent locomotive road inspection and detection system
TWI803787B (en) * 2020-11-05 2023-06-01 中華電信股份有限公司 Method and system for classification and identification of road defilement

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI803787B (en) * 2020-11-05 2023-06-01 中華電信股份有限公司 Method and system for classification and identification of road defilement
TWI765558B (en) * 2021-02-02 2022-05-21 覺華工程科技股份有限公司 Intelligent locomotive road inspection and detection system

Similar Documents

Publication Publication Date Title
CN108318043B (en) Method, apparatus, and computer-readable storage medium for updating electronic map
CN109084786B (en) Map data processing method
CN101694084B (en) Ground on-vehicle mobile detecting system
CN113424232A (en) Three-dimensional point cloud map construction method, system and equipment
JP5404861B2 (en) Stationary object map generator
JP5388082B2 (en) Stationary object map generator
CN111006655A (en) Multi-scene autonomous navigation positioning method for airport inspection robot
CN112380312B (en) Laser map updating method based on grid detection, terminal and computer equipment
CN112455502B (en) Train positioning method and device based on laser radar
CN112308913B (en) Vehicle positioning method and device based on vision and vehicle-mounted terminal
US20200279395A1 (en) Method and system for enhanced sensing capabilities for vehicles
TWM600873U (en) Detection system for detecting road damage
CN103185578A (en) Mobile positioning device and navigation device
CN110018503B (en) Vehicle positioning method and positioning system
Abbate et al. Prospective upon multi-source urban scale data for 3d documentation and monitoring of urban legacies
CN115564865A (en) Construction method and system of crowdsourcing high-precision map, electronic equipment and vehicle
Aggarwal GPS-based localization of autonomous vehicles
Kaya et al. Car detection from very high-resolution UAV images using deep learning algorithms
TW202136752A (en) Detection system and method for detecting road damage distinguishing road damage according to the image and calculating and storing instantaneous coordinates of the GNSS-RTK moving device according to the observation data
CN115409910A (en) Semantic map construction method, visual positioning method and related equipment
Luo et al. Automatic mileage positioning for road inspection using binocular stereo vision system and global navigation satellite system
Koppanyi et al. Experiences with acquiring highly redundant spatial data to support driverless vehicle technologies
CN111652276A (en) All-weather portable multifunctional bionic positioning, attitude determining and viewing system and method
WO2023222671A1 (en) Position determination of a vehicle using image segmentations
US20220404170A1 (en) Apparatus, method, and computer program for updating map