100年09月19日修正替換π M420791 五、新型說明： 【新型所屬之技術領域】 [0001] 本申請是有關於一種車輛行駛違規車道的自動化偵測方 法與裝置，且特別是有關於一種測定一或多耗車是否違 規行駛車道以及利用影像處理技術來自動偵測交通違規 車輛之方法與裝置。 【先前技術】Correction and replacement of π M420791 on September 19, 100. V. New description: [New technical field] [0001] The present application relates to an automatic detection method and device for driving lanes of vehicles, and in particular to a measurement One or more vehicles are illegally driving lanes and methods and devices for automatically detecting traffic violation vehicles using image processing technology. [Prior Art]
[0002] 目前車輛行駛違規車道取締十分困難，由於車輛在行駛 中，取締不易且有其危險性。可能的作法是以人工或自 動照相取得正在違規之車輛影像，在加以人工判定是否 違規舉發。 【新型内容】 [0003] 為一種新型車輛違規行駛車道自動偵測系統；在有限制 車道路口（如機車專用道、汽車專用道、公車專用道或高 速公路内側車道等），裝設攝影裝置，利用所得影像及影 像處理技術，在系統架設時完成道路環境辨識（道路標線 、標誌、道路地面文字等），產生車道行駛限制；然後再 利用在限制車道道路攝影點上所得即時影像，完成車牌 辨識及車輛型別辨識等功能，計算得到車輛車牌資訊、 車輛型別及其行駛之車道。如果該車輛不應行駛在該限 制車道，則該車輛已違規行驶車道；本系統便可以攝影 點上所得影像及所辨識出的車牌號碼，告發該車輛違規 行駛車道。本系統相較於先前技術，具有自動化、可執 行性及減少執行時人員之危險等功效，且具有產業利用 性、新穎性及進步性。 表單編號Α0101 第3頁/共9頁 100年09月19日核正替換哀 【實施方式】 ---- _現在將舉其較佳實施例，並參照所附圖式，予以更完整 地說明》 第1圖是根據本新型之數種較佳實施例（（a)機車專用道 、⑴禁行機車it、（e)高速高路_車道）及系統方塊 參照第㈣，上述車輛違規變換車道自動偵測裝置包括攝 影裝置100、網路單元110、道路環境辨識單元12〇、車 牌車種辨識單元130、車輛違規行歇車道辨識單元140、 儲存單元150以及前端顯示單元16〇。 參照第1圖，攝影裝置1〇〇被安裝在限制行驶車道的道路 上（如機車專用道、汽車專用道、公車專用道或高速公路 内側車道等）以攝影偵_動中的車輛。在系統架設完成 之時，道路環境辨識單元120能以所得之道路影像，自動 辨識為-種新型車輛違規行驶車道自動傾測系統；在有 限制車道路口（如機車專用道、汽車專用道、公車專用道 或高速公路内側車道等），裝設攝影裝置，利用所得影像 及影像處理技術，在系統架設時完成道路環境辨識（道路 標線、標誌、道路地面文字等），產生車道行駛限制；然 後再利用在限制車道道路攝影點上所得即時影像，完成 車牌辨識及車輛型別辨識等功能，計算得到車輛車牌資 訊、車輛型別及其行驶之車道。如果該車輛不應行駛在 該限制車道，則該車輛已違規行駛車道；本系統便可以 攝影點上所得影像及所辨識出的車牌號瑪，告發該車輛 違規行駛車道《本系統相較於先前技術，具有自動化、 可執行性及減少執行時人員之危險等功效。，並產生該 第4頁/共9頁 表單編號A0101 地點車道行勒 f 100年.09月19.日# 當車輛通過攝二人機介面輪"行駛， 理單元至影像處 車輛違規料車道_以_轉㈣賴R130及 車耗型別（車種X大客貨車)客’；；辨識其車牌號瑪、 敬車道。此吐^ J吝貝車、機慢車等）及行 述所辨識之車牌：=行歇車道辨識單元140將根據上 等決定#、行駛車道及車道行驶規則 觀料車道； 及相關資料儲存在儲存單元⑽淮供已延規則將衫像 相關f # Μ _ 準備告發；或將影像及 ㈣經網路110傳至告發單位。 及乘宏，喊仃驶車輯識單元140可⑽識前座駕駛 疋否沒繫女全帶（當攝影裝置自車輛前端攝影時） 單元15=_將違規車_像及相關資_存在儲存 準備告發，或將影像及相關資料經網路110傳 至告發單位。 、統車輛違規行駛車道辨識單元14Gm對經過攝影 點之車辆車牌號碼與重要追縱車輛資料（如賊車、重大案 件嫌疑車、通缉車等如發現符合，將違規將影像及相 關資u網路11Q傳至攝影點之後之攔檢點（前端顯示單 元160)。 本系統之道路環境辨識單元120，係以影像處理技術，在 系統架設完成之時’先建構道路背景，再利用特徵淬取 及樣板比對’自動辨識道路標線、標誌、道路地面文字 等’並產生該地點車道行駛規則；亦可以人機介面輸入行 駛規則。 本系統之車牌車種辨識單元130，係以影像處理技術，先 表單編號A0101 第5頁/共9頁 M420791 100年09月19日修正替換苜 在影像中找出可能車牌位置，再以字元辨識技術確定車 牌號碼。車種辨識則先自影像晝面學習各車種差異，再 利用此結果判定後續影像中之車輛種類。 第1圖係以平面圖顯示本申請之數個實施例及系統方塊圖[0002] At present, it is very difficult to ban the violation lane of a vehicle, and it is difficult and dangerous to ban because the vehicle is running. It is possible to obtain images of vehicles that are being violated by manual or automatic photography, and to manually determine whether or not to violate the rules. [New content] [0003] is a new type of vehicle violation lane automatic detection system; in the limited vehicle road (such as locomotive lanes, car lanes, bus lanes or highway lanes, etc.), equipped with photographic equipment, Using the obtained image and image processing technology, road environment identification (road markings, signs, road ground text, etc.) is completed during system erection, and lane driving restrictions are generated; then, the real-time images obtained at the restricted lane road photography points are used to complete the license plate. Identification and vehicle type identification, etc., calculate vehicle license plate information, vehicle type and driving lane. If the vehicle is not allowed to drive in the restricted lane, the vehicle has violated the driving lane; the system can photograph the obtained image and the identified license plate number to inform the vehicle of the illegal driving lane. Compared with the prior art, the system has the functions of automation, executability and reduction of the danger of personnel during execution, and has industrial utilization, novelty and progress. Form No. 1010101 Page 3 of 9 Page 19 September 19th Nuclear Replacement [Embodiment] ---- _ The preferred embodiment will now be described, and will be more fully described with reference to the accompanying drawings. 1 is a plurality of preferred embodiments according to the present invention ((a) locomotive lane, (1) forbidden locomotive it, (e) high speed highway _ lane) and system block reference (4), the above vehicle violates the lane change lane The automatic detecting device includes a photographing device 100, a network unit 110, a road environment recognizing unit 12, a license plate type identifying unit 130, a vehicle illegal lane identifying unit 140, a storage unit 150, and a front end display unit 16A. Referring to Fig. 1, the photographing apparatus 1 is mounted on a road that restricts the traveling lane (e.g., a locomotive lane, a car lane, a bus lane, or a highway lane, etc.) to detect a moving vehicle. When the system is set up, the road environment identification unit 120 can automatically recognize the road image as a new type of vehicle illegal driving lane automatic tilting detection system; in the limited road intersection (such as the locomotive lane, the automobile lane, the bus) a dedicated road or a highway inside lane, etc., equipped with a photographic device, using the obtained image and image processing technology to complete road environment identification (road markings, signs, road ground text, etc.) during system erection, and generate lane driving restrictions; The real-time image obtained on the road marking point of the restricted lane is used to complete the functions of license plate recognition and vehicle type identification, and the vehicle license plate information, the vehicle type and the driving lane are calculated. If the vehicle should not drive in the restricted lane, the vehicle has violated the driving lane; the system can photograph the obtained image and the identified license plate number, and report the illegal driving lane of the vehicle. Technology, with automation, enforceability and reduced risk of personnel during execution. And generate the 4th page / a total of 9 pages form number A0101 location lane line le f 100 years. September 19. day # When the vehicle passes the camera two-machine interface wheel " travel, the unit to the image vehicle violation lane _ _ turn (four) 赖R130 and car consumption type (vehicle type X big van) passenger;; identify its license plate number Ma, Jing Lane. The vehicle license plate identified by the vehicle, the vehicle, and the vehicle: the lane recognition unit 140 will store the traffic lane according to the superior decision #, the driving lane and the lane driving rule; and the related data are stored in the storage Unit (10) Huaiwei has extended the rules to tie the shirt like f # Μ _ to prepare for the announcement; or to transmit the image and (4) to the reporting unit via the network 110. And take the macro, shouting the car collection unit 140 can (10) know the front seat driving, whether the female full belt (when the camera is photographed from the front of the vehicle) unit 15 = _ will be illegal car _ image and related resources _ stored preparation Report or transmit images and related materials to the reporting unit via the Internet 110. The illegal vehicle lane recognition unit 14Gm will violate the regulations and the relevant information if the vehicle license plate number and the important vehicle tracking information (such as the thief car, the major case suspect car, the wanted car, etc.) are met. The road 11Q is transmitted to the checkpoint after the photographing point (front end display unit 160). The road environment recognizing unit 120 of the system adopts image processing technology to construct the road background first, and then utilize feature extraction and when the system is completed. The model compares 'automatically identify road markings, signs, road ground texts, etc.' and generates lane driving rules for the location; it can also input driving rules by human-machine interface. The license plate type identification unit 130 of the system is based on image processing technology. Form No. A0101 Page 5 of 9 M420791 Correction of replacement on September 19, 100. Find the possible license plate position in the image, and then use the character recognition technology to determine the license plate number. The vehicle type identification first learns the various types of vehicles from the image. Difference, and use this result to determine the type of vehicle in the subsequent image. FIG. 1 shows a plurality of embodiments of the present application in a plan view. System block diagram
雖然本申請已經以其較佳實施例揭露如上，然其並非用 以限定本申請，任何熟習此技藝者，在不脫離本申請之 精神的情況下，當可作些許之更動，因此本申請之權利 保護範圍當視後附之申請專利範圍所界定者為準。 【圖式簡單說明】 [0005] 第1圖係以平面圖顯示本申請之較佳實施例及系統組態方 塊圖。 【主要元件符號說明】 [0006] 100 · · ·攝影裝置 11 0 · · ·網路單元 120 ···道路環境辨識單元 130 ···車牌車種辨識單元 140· ··車輛違規行駛車道辨識單元 150 · · ·儲存單元 160 ···前端顯示單元 表單編號A0101 第6頁/共9頁Although the present application has been disclosed in its preferred embodiments as above, it is not intended to limit the application, and any person skilled in the art can make some changes without departing from the spirit of the application. The scope of protection of the rights is subject to the definition of the scope of the patent application. BRIEF DESCRIPTION OF THE DRAWINGS [0005] Fig. 1 is a plan view showing a preferred embodiment of the present application and a system configuration block diagram. [Description of main component symbols] [0006] 100 · · · Photography device 11 0 · · Network unit 120 ··· Road environment identification unit 130 ··· License plate type identification unit 140 · · Vehicle violation driving lane recognition unit 150 · · · Storage unit 160 ··· Front display unit form number A0101 Page 6 of 9