TWI798022B

TWI798022B - A reminder method and system for road indicating objects

Info

Publication number: TWI798022B
Application number: TW111108770A
Authority: TW
Inventors: 王冠傑; 呂京霖; 何宗璟
Original assignee: 台灣智慧駕駛股份有限公司
Priority date: 2022-03-10
Filing date: 2022-03-10
Publication date: 2023-04-01
Also published as: TW202335894A

Abstract

一種道路指示物件的提醒方法及其系統，應用於行駛在一私人道路的一車輛，該方法包括以下步驟：先取得至少一車輛週邊影像，然後對該車輛週邊影像進行一影像特徵萃取，萃取至少一關聯該私人道路上的至少一道路指示物件的特徵，接著，對該特徵進行一座標轉換，以取得該道路指示物件相對於該車輛的一相對位置，再根據該特徵以及該相對位置對該道路指示物件進行辨識，判斷是否符合複數個預設條件的至少其中一者，以產生一辨識結果，最後，響應該辨識結果在該車輛執行一操作。該方法可不使用一地圖資訊及該車輛的一定位資訊，增加應用上的彈性。A method and system for reminding road indication objects, applied to a vehicle driving on a private road, the method includes the following steps: first obtain at least one surrounding image of the vehicle, and then perform image feature extraction on the surrounding image of the vehicle, extracting at least A feature associated with at least one road indicating object on the private road, and then performing coordinate conversion on the feature to obtain a relative position of the road indicating object relative to the vehicle, and then according to the feature and the relative position to the The road indicating object is identified, and it is judged whether at least one of a plurality of preset conditions is met to generate an identification result, and finally, an operation is performed on the vehicle in response to the identification result. The method does not use a map information and a location information of the vehicle, which increases the flexibility of application.

Description

一種道路指示物件的提醒方法及其系統Reminder method and system for road indication objects

本發明是有關於一種控制運輸載具的輔助方法及其系統，且特別關於一種道路指示物件的提醒方法及其系統。 The present invention relates to an auxiliary method and system for controlling a transport vehicle, and in particular to a reminding method and system for road indicating objects.

現今的車輛大多搭載有安全駕駛系統，以提升行車安全，並降低行車意外或事故的可能。最廣為應用的為先進輔助駕駛系統(Advanced Driver Assistance Systems，ADAS)，其中，影像辨識為該系統中的一重要技術。透過車用攝影機取得車輛周圍的環境影像，經過影像處理以及識別後，幫助駕駛人注意周遭的行車動態，提升安全性以及便利性。 Most of today's vehicles are equipped with safety driving systems to improve driving safety and reduce the possibility of driving accidents or accidents. The most widely used is Advanced Driver Assistance Systems (ADAS), wherein image recognition is an important technology in the system. The environment image around the vehicle is obtained through the vehicle camera. After image processing and recognition, it helps the driver to pay attention to the driving dynamics around him, improving safety and convenience.

對於車用影像辨識來說，大多利用車載定位裝置提供的車輛位置資訊，配合一般地圖或高精地圖，提供使用者有關附近道路標示的資訊。例如利用識別道路標示輔助駕駛人停車(例如US10235580B2)或提供駕駛人關於道路周遭告示牌(例如US10713510B2)或標示的資訊(例如US10984551B2)。 For vehicle image recognition, the vehicle location information provided by the vehicle positioning device is mostly used together with general maps or high-definition maps to provide users with information about nearby road signs. For example, the recognition of road signs is used to assist the driver to park (such as US10235580B2) or to provide the driver with information about signs (such as US10713510B2) or signs around the road (such as US10984551B2).

然而，在現有方法中，大多需要配合地圖資訊，因而在應用上受到了限制；另一方面，也由於需要配合地圖資訊，對於私人場域來說，例如工廠、校園、遊樂園、高爾夫球場或社區等場域，其道路配置或道路標示的資訊不一定會同步地顯示於地圖資訊，因此，並無法滿足在私人場域的應用。 However, in the existing methods, most of them need to cooperate with map information, so the application is limited; on the other hand, due to the need to cooperate with map information, for private fields such as factories, campuses, amusement parks, golf courses or In areas such as communities, the road configuration or road marking information may not be displayed on the map information synchronously. Therefore, it cannot meet the application in private areas.

本發明提供一種道路指示物件的提醒方法及其系統，適用於私人場域，能在不使用地圖資訊或定位資訊的情況下，對車輛周遭的道路指示物件進行辨識以及提醒。 The present invention provides a method and system for reminding road indicating objects, which are suitable for private fields, and can recognize and remind road indicating objects around vehicles without using map information or positioning information.

本發明一實施例提供一種道路指示物件的提醒方法，包括以下步驟：步驟1：取得行駛在該私人道路上的該車輛的至少一車輛週邊影像；步驟2：對該車輛週邊影像進行一影像特徵萃取，以萃取出至少一關聯該私人道路上的至少一道路指示物件的特徵，該特徵至少包括該道路指示物件的一輪廓、一顏色、一文字、一符號、一圖像、一尺寸、一位置或前述的組合；步驟3：對該特徵進行一座標轉換，以取得該道路指示物件相對於該車輛的一相對位置；步驟4：根據該特徵以及該相對位置對該道路指示物件進行辨識，判斷是否符合一資料庫中的複數個預設條件的至少其中一者，以產生一辨識結果，其中，該資料庫的該預設條件是根據該私人道路的該道路指示物件而預先定義；以及步驟5：響應該辨識結果在該車輛執行一操作。 An embodiment of the present invention provides a method for reminding road indication objects, including the following steps: Step 1: Obtain at least one vehicle surrounding image of the vehicle driving on the private road; Step 2: Perform an image feature on the vehicle surrounding image Extracting to extract at least one feature associated with at least one road indicating object on the private road, the feature at least including an outline, a color, a text, a symbol, an image, a size, and a position of the road indicating object Or a combination of the foregoing; Step 3: Carry out a coordinate transformation on the feature to obtain a relative position of the road indicating object relative to the vehicle; Step 4: Identify and judge the road indicating object according to the feature and the relative position whether at least one of a plurality of preset conditions in a database is met to generate a recognition result, wherein the default condition of the database is predefined according to the road indicator object of the private road; and steps 5: Perform an operation on the vehicle in response to the identification result.

本發明另一實施例提供一種道路指示物件的提醒系統，包括：至少一影像擷取裝置，安裝在一車輛上，該影像擷取裝置取得該車輛行駛在一私人道路上的車輛的至少一車前影像；以及一處理單元，耦接至該影像擷取裝置；其中，該處理單元被配置為執行以下步驟：對該車輛週邊影像進行一影像特徵萃取，以萃取出至少一關聯該私人道路上的至少一道路指示物件的特徵，該特徵至少包括該道路指示物件的一輪廓、一顏色、一文字、一符號、一圖像、一尺寸、一位置或前述的組合；對該特徵進行一座標轉換，以取得該道路指示物件相對於該車輛的一相對位置；根據該特徵以及該相對位置對該道路指示物件進行辨識，判斷是否符合一資料庫中的複數個預設條件的至少其中一者，以產生一辨識結果，其中，該資料庫的該預設條件是根據該私人道路的該道路指示物件而預先定義；以及響應該辨識結果在該車輛執行一操作。 Another embodiment of the present invention provides a reminder system for road indicator objects, including: at least one image capture device installed on a vehicle, the image capture device captures at least one vehicle of the vehicle driving on a private road a front image; and a processing unit coupled to the image capture device; wherein the processing unit is configured to perform the following steps: performing an image feature extraction on the surrounding image of the vehicle to extract at least one feature related to at least one road indicating object on the private road, the feature at least including an outline, a color, a character, a symbol, An image, a size, a position, or a combination of the foregoing; performing coordinate conversion on the feature to obtain a relative position of the road indicating object relative to the vehicle; performing an operation on the road indicating object according to the feature and the relative position Identifying, judging whether at least one of a plurality of preset conditions in a database is met, so as to generate a recognition result, wherein the default condition in the database is predefined according to the road indication object of the private road ; and performing an operation on the vehicle in response to the identification result.

10:系統 10: System

11:處理單元 11: Processing unit

12:影像擷取裝置 12: Image capture device

13:輸出單元 13: Output unit

20A:房車 20A: RV

20B:高爾夫球車 20B: Golf Cart

21:駕駛艙 21: Cockpit

22:頂篷 22: Canopy

30:私人道路 30: Private Road

40:車輛週邊影像 40:Vehicle surrounding image

50A、50B:交通標線、第一交通標線、第二交通標線 50A, 50B: traffic markings, first traffic markings, second traffic markings

50C:交通標誌 50C: Traffic signs

50D:交通號誌 50D: Traffic signal

S0、S1、S2、S3、S4、S5:步驟 S0, S1, S2, S3, S4, S5: steps

『圖1A』至『圖1B』分別是根據本發明一實施例的系統應用於車輛的示意圖。 [FIG. 1A] to [FIG. 1B] are schematic diagrams of a system applied to a vehicle according to an embodiment of the present invention.

『圖2』是根據本發明一實施例的提醒方法的步驟流程圖。 [FIG. 2] is a flowchart of steps of a reminder method according to an embodiment of the present invention.

『圖3A』至『圖3D』說明本發明不同實施例的車輛週邊影像的示意圖。 [FIG. 3A] to [FIG. 3D] illustrate schematic diagrams of vehicle surrounding images in different embodiments of the present invention.

在本文中，儘管一些實施例以特定順序來執行步驟，但是仍可以另一合理之順序來執行這些步驟。對於不同的實施例，以下說明的某些特徵可以替換或消除。應理解的是，一些額外的操作可以在所敘述的方法之前、之中或之後執行，並且在該方法的其他實施例中，某些操作可以被替換或省略。 Herein, although some embodiments perform steps in a particular order, the steps may be performed in another reasonable order. Certain features described below may be substituted or eliminated for different embodiments. It should be understood that some additional operations may be performed between the described methods performed before, during, or after, and in other embodiments of the method, certain operations may be replaced or omitted.

在本文中，對各種實施例的描述中所使用的術語只是為了描述特定示例的目的，而並非旨在進行限制。除非上下文另外明確地表明，或刻意限定元件的數量，否則本文所用的單數形式“一”、“一個”及“該”也包含複數形式。 The terminology used in describing various embodiments herein is for the purpose of describing particular examples only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" also include the plural forms unless the context clearly dictates otherwise, or the number of elements is intentionally limited.

『圖1A』至『圖1B』是根據本發明一實施例的系統應用於車輛的示意圖，該系統10包括一處理單元11、至少一影像擷取裝置12及一輸出單元13，該系統10安裝在一車輛上。在『圖1A』的實施例中，該車輛為一房車20A，該影像擷取裝置12安裝在該房車20A的一駕駛艙21的前方，而該處理單元11裝設於該房車20A中，該輸出單元13設置於該駕駛艙21內；在『圖1B』的實施例中，該車輛為一高爾夫球車20B，該影像擷取裝置12安裝在該高爾夫球車20B的一頂篷22，而該處理單元11裝設於該高爾夫球車20B中，該輸出單元13設置於該高爾夫球車20B的一駕駛艙21內。根據本發明一實施例，該輸出單元13可以是一顯示裝置或一揚聲器，且該輸出單元13可以是固定於該車輛上的裝置，也可以是一分離於該車輛上的裝置，例如配戴於駕駛人上的穿戴式電子裝置或手持式電子裝置。 "Fig. 1A" to "Fig. 1B" are schematic diagrams of a system applied to a vehicle according to an embodiment of the present invention. The system 10 includes a processing unit 11, at least one image capture device 12 and an output unit 13. The system 10 is installed on a vehicle. In the embodiment of "Fig. 1A", the vehicle is a caravan 20A, the image capture device 12 is installed in front of a cockpit 21 of the caravan 20A, and the processing unit 11 is installed in the caravan 20A, the The output unit 13 is arranged in the cockpit 21; in the embodiment of "Fig. 1B", the vehicle is a golf cart 20B, the image capture device 12 is installed on a roof 22 of the golf cart 20B, and The processing unit 11 is installed in the golf cart 20B, and the output unit 13 is arranged in a driving compartment 21 of the golf cart 20B. According to an embodiment of the present invention, the output unit 13 can be a display device or a speaker, and the output unit 13 can be a device fixed on the vehicle, or a device separated from the vehicle, such as a wearable Wearable electronic devices or handheld electronic devices on the driver.

本發明是應用在行駛在一私人道路的一車輛，該私人道路係指非公共道路的路面，特別是指道路指示物件不符合政府或交通管理單位規範的道路。本發明中，道路指示物件包括但不限於交通標誌(Traffic sign)、交通標線(Road surface marking)、交通號誌(Traffic light)等用於提供道路或行駛資訊的可動或固定式裝置。 The present invention is applied to a vehicle running on a private road. The private road refers to the road surface of a non-public road, especially a road whose road indicator objects do not conform to the regulations of the government or traffic management unit. In the present invention, road indicating objects include but not limited to traffic signs, road surface markings, traffic lights and other movable or fixed devices for providing road or driving information.

『圖2』是根據本發明一實施例的方法流程圖，各個步驟的說明如下。 [FIG. 2] is a flowchart of a method according to an embodiment of the present invention, and each step is described as follows.

在該車輛行駛前，先將上述的道路指示物件輸入一資料庫中，並對該資料庫進行訓練，以建立一之後用於影像辨識的模型(步驟S0)。 Before the vehicle runs, the above-mentioned road indication objects are input into a database, and the database is trained to establish a model for image recognition (step S0).

當該車輛在一私人道路30行駛時，利用該影像擷取裝置12取得該車輛的至少一車輛週邊影像40(步驟S1)，該車輛週邊影像40如『圖3A』所示，該車輛週邊影像40是利用該影像擷取裝置12所取得從該車輛所觀看的一環境的一部分。本實施例中，該車輛週邊影像40至少包括一車前影像，該部分包括該車輛所行駛的一路面以及鄰接於該路面的一部分。在其他實施例中，該車輛週邊影像40還可以進一步包括該車輛的一左側影像、一右側影像、一車後影像或前述的組合，也可以是例如從該車輛面向其他方位的影像。 When the vehicle is driving on a private road 30, use the image capture device 12 to obtain at least one vehicle surrounding image 40 of the vehicle (step S1), the vehicle surrounding image 40 is shown in "Fig. 3A", the vehicle surrounding image 40 is a part of an environment viewed from the vehicle obtained by the image capture device 12 . In this embodiment, the surrounding image 40 of the vehicle at least includes a front image, and the part includes a road surface on which the vehicle is driving and a part adjacent to the road surface. In other embodiments, the vehicle peripheral image 40 may further include a left side image, a right side image, a rear image of the vehicle or a combination of the foregoing, and may also be, for example, an image from the vehicle in other directions.

接著，對該車輛週邊影像40進行一影像特徵萃取(Image feature extraction)，以定義出至少一關聯該私人道路上的至少一該道路指示物件的特徵，該特徵至少包括該道路指示物件的一輪廓、一顏色、一文字、一符號、一圖像、一尺寸、一位置、一設置角度或前述的組合(步驟S2)。 Next, an image feature extraction (Image feature extraction) is performed on the surrounding image 40 of the vehicle to define at least one feature associated with at least one road indicator object on the private road, the feature at least including a contour of the road indicator object , a color, a character, a symbol, an image, a size, a position, a setting angle or a combination thereof (step S2).

本實施例中，該道路指示物件為一交通標線50A，該交通標線50A繪製於該私人道路30的一右側，該交通標線50A為一具有一第一顏色的實線，如『圖3A』所示。但本發明並不限於此，在其他實施例中，該道路指示物件可包括一第一交通標線50A及一第二交通標線50B，該第一交通標線50A及該第二交通標線50B分別為一第一顏色及一第二顏色，如『圖3B』所示。該第一顏色及該第二顏色可為不同或相同的顏色，以表示相同或不同的資訊，該資訊例如禁止行駛路段、危險路段、安全路段、慢速行駛路段、注意路面陡峭、禁止進入、禁止會車、路側陡峭或小心駕駛路段等。此外，在其他實施例中，該道路指示物件可為一交通標誌50C，如『圖3C』所示，或者，該道路指示物件可為一交通號誌50D，如『圖3D』所示(根據標示角度預判曲率)。該影像特徵萃取係萃取該車輛週邊影像40中的該道路指示物件的該特徵，該影像特徵萃取可採用一神經網路法、一影像梯度法或一聚類法等。待萃取該車輛週邊影像40中的該道路指示物件的該特徵後，對該特徵進行一座標轉換，以取得該道路指示物件相對於該車輛的一相對位置(步驟S3)。在進行該座標轉換前，根據該車輛週邊影像40所得到的為該特徵相對於該影像擷取裝置12在空間中的一相對位置(即該特徵在該車輛週邊影像40中的相對位置)。在進行該座標轉換後，可得到該特徵相對該車輛在空間中的一相對位置。該座標轉換可以利用一逆透視投影法、一數學幾何法或一光達點雲萃取法來進行。 In this embodiment, the road indicating object is a traffic marking 50A, which is drawn on the right side of the private road 30, and the traffic marking 50A is a solid line with a first color, as shown in Fig. 3A". But the present invention is not limited thereto. In other embodiments, the road indicating object may include a first traffic marking 50A and a second traffic marking 50B, and the first traffic marking 50A and the second traffic marking 50B are respectively a first color and a second color, as shown in "Fig. 3B". The first color and the second color can be different or the same color to represent the same or different information, such as forbidden road section, dangerous road section, safe road section, slow speed section, Pay attention to steep roads, no entry, no passing cars, steep roadsides, or drive carefully on road sections, etc. In addition, in other embodiments, the road indicating object can be a traffic sign 50C, as shown in "Fig. 3C", or, the road indicating object can be a traffic signal 50D, as shown in "Fig. 3D" (according to Mark the angle to predict the curvature). The image feature extraction is to extract the feature of the road indicator object in the vehicle peripheral image 40 , and the image feature extraction can use a neural network method, an image gradient method, or a clustering method. After the feature of the road indicator object in the vehicle peripheral image 40 is extracted, coordinate transformation is performed on the feature to obtain a relative position of the road indicator object relative to the vehicle (step S3 ). Before the coordinate transformation, a relative position of the feature in space relative to the image capture device 12 (ie, the relative position of the feature in the vehicle surrounding image 40 ) is obtained according to the vehicle surrounding image 40 . After the coordinate transformation is performed, a relative position of the feature relative to the vehicle in space can be obtained. The coordinate transformation can be performed by using an inverse perspective projection method, a mathematical geometry method or a lidar point cloud extraction method.

本實施例中，該影像特徵萃取及該座標轉換是透過該處理單元11或經由該處理單元11連接至一遠端伺服器來執行。此外，該影像特徵萃取及該座標轉換並不使用一地圖資訊及該車輛的一定位資訊，僅需要該影像擷取裝置12以及執行影像辨識的該處理單元11即可完成，可以大幅簡化硬體配置。 In this embodiment, the image feature extraction and the coordinate conversion are performed through the processing unit 11 or connected to a remote server through the processing unit 11 . In addition, the image feature extraction and the coordinate conversion do not use a map information and a location information of the vehicle, and only need the image capture device 12 and the processing unit 11 for image recognition to complete, which can greatly simplify hardware configuration.

完成該座標轉換後，根據該特徵以及該相對位置對該道路指示物件進行辨識，判斷是否符合該資料庫中的複數個預設條件的至少其中一者，以產生一辨識結果(步驟S4)。該資料庫可儲存於裝設於該車輛的一記憶單元或位於該遠端伺服器的一儲存單元。 After the coordinate transformation is completed, the road indicator object is identified according to the feature and the relative position, and whether at least one of a plurality of preset conditions in the database is met is judged to generate an identification result (step S4). The database can be stored in a memory unit installed in the vehicle or a storage unit in the remote server.

該處理單元11判斷該道路指示物件的該特徵是否符合該些預設條件中的部分項目。舉例來說，該處理單元11先決定該道路指示物件是否符合交通標誌、交通標線或交通號誌中的任一物件。若符合，該處理單元11再進一步判斷該物件所表示的資訊，該處理單元11根據該資訊而產生一第一辨識結果。若不符合，該處理單元11產生一第二辨識結果。 The processing unit 11 judges whether the feature of the road indication object meets some of the preset conditions. For example, the processing unit 11 first determines whether the road indication object meets Any object in a traffic sign, traffic marking, or traffic signal. If yes, the processing unit 11 further judges the information represented by the object, and the processing unit 11 generates a first recognition result according to the information. If not, the processing unit 11 generates a second identification result.

當判斷後，響應該辨識結果在該車輛執行一操作(步驟S5)。根據本發明一實施例，若判斷後產生該第一辨識結果，表示該車輛的駕駛人應注意附近的該道路指示物件，該操作可以是由該處理單元11發送一警示訊號給該輸出單元13，由該輸出單元13對該車輛的駕駛人發出一提醒，該提醒乃關聯於該第一辨識結果，該提醒可以是一視覺通知、一聲音通知、一觸覺回饋或前述的組合；根據本發明另一實施例，該操作也可以是直接對該車輛進行一控制。若判斷後產生該第二辨識結果，表示該車輛的附近不存在該車輛的駕駛人須注意的物件，將不執行該操作。 After judging, perform an operation on the vehicle in response to the recognition result (step S5). According to an embodiment of the present invention, if the first recognition result is generated after the judgment, it means that the driver of the vehicle should pay attention to the nearby road indicating object, and the operation may be that the processing unit 11 sends a warning signal to the output unit 13 , the output unit 13 sends a reminder to the driver of the vehicle, the reminder is related to the first identification result, the reminder can be a visual notification, a sound notification, a tactile feedback or a combination of the foregoing; according to the present invention In another embodiment, the operation may also be to directly control the vehicle. If the second recognition result is generated after the judgment, it means that there is no object near the vehicle that the driver of the vehicle should pay attention to, and the operation will not be executed.

在一例子中，當應用在一彎曲路段，且該彎曲路段設置有一指示該彎曲路段的交通標誌時，該影像特徵萃取是判斷該道路指示物件設置在該彎曲路段的該設置角度(即該特徵)，並判斷該設置角度是否符合該資料庫中的複數個預設條件的至少其中一者，藉此預判該車輛前方的該彎曲路段的一曲率。本發明揭露的方法以及系統，並不須使用一地圖資訊及該車輛的一定位資訊，大幅降低硬體配置的需求。也由於毋須使用地圖資訊，該車輛可不搭載如GPS的定位模組，適合用在訊號不佳的地區，另一方面，由於毋須使用地圖資訊，故可不依靠地圖供應商提供的一般地圖或高精地圖，也不用預先建置應用場域的地貌或地圖，只要預先對該道路指示物件進行建模，配合影像辨識即可使用，適合用在私人場域中，應用彈性大。 In one example, when it is applied to a curved road section, and the curved road section is provided with a traffic sign indicating the curved road section, the image feature extraction is to determine the installation angle of the road indicating object on the curved road section (that is, the feature ), and determine whether the setting angle meets at least one of a plurality of preset conditions in the database, so as to predict a curvature of the curved road section ahead of the vehicle. The method and system disclosed in the present invention do not need to use a map information and a location information of the vehicle, which greatly reduces the requirement of hardware configuration. Also because there is no need to use map information, the vehicle does not need to be equipped with a positioning module such as GPS, which is suitable for use in areas with poor signal. On the other hand, because there is no need to use map information, it does not need to rely on general maps or high-precision The map does not need to pre-build the topography or map of the application field, as long as the road indication object is modeled in advance, and it can be used with image recognition. It is suitable for use in private fields and has great application flexibility.

S0、S1、S2、S3、S4、S5:步驟 S0, S1, S2, S3, S4, S5: steps

Claims

一種道路指示物件的提醒方法，應用於行駛在一私人道路的一車輛，該方法包括以下步驟：步驟1：取得行駛在該私人道路上的該車輛的至少一車輛週邊影像；步驟2：對該車輛週邊影像進行一影像特徵萃取，以萃取出至少一關聯該私人道路上的至少一道路指示物件的特徵，該特徵至少包括該道路指示物件的一輪廓、一顏色、一文字、一符號、一圖像、一尺寸、一位置或前述的組合；步驟3：對該特徵進行一座標轉換，以取得該道路指示物件相對於該車輛的一相對位置；步驟4：根據該特徵以及該相對位置對該道路指示物件進行辨識，判斷是否符合一資料庫中的複數個預設條件的至少其中一者，以產生一辨識結果，其中，該資料庫的該預設條件是根據該私人道路的該道路指示物件而預先定義；以及步驟5：響應該辨識結果在該車輛執行一操作。 A method for reminding a road indicating object, which is applied to a vehicle driving on a private road, the method includes the following steps: Step 1: Obtain at least one vehicle peripheral image of the vehicle driving on the private road; Step 2: An image feature extraction is performed on the surrounding image of the vehicle to extract at least one feature related to at least one road indicating object on the private road, and the feature includes at least an outline, a color, a character, a symbol, and a picture of the road indicating object image, a size, a position or a combination of the foregoing; step 3: perform coordinate conversion on the feature to obtain a relative position of the road indicating object relative to the vehicle; step 4: according to the feature and the relative position to the Identifying road indication objects and judging whether they meet at least one of a plurality of preset conditions in a database to generate a recognition result, wherein the preset conditions in the database are based on the road indication of the private road object; and step 5: performing an operation on the vehicle in response to the recognition result.

如請求項1的方法，其中在執行該方法時，不使用一地圖資訊及/或該車輛的一定位資訊。 The method according to claim 1, wherein when performing the method, map information and/or location information of the vehicle are not used.

如請求項1的方法，其中該操作是對該車輛的一操作者發出一關聯於該辨識結果的提醒，該提醒為一視覺通知、一聲音通知、一觸覺回饋或前述的組合。 The method according to claim 1, wherein the operation is to issue a reminder related to the recognition result to an operator of the vehicle, and the reminder is a visual notification, an audio notification, a tactile feedback or a combination thereof.

如請求項1的方法，其中該操作是對該車輛進行一控制。 The method of claim 1, wherein the operation is a control of the vehicle.

一種道路指示物件的提醒系統，包括：至少一影像擷取裝置，安裝在一車輛上，該影像擷取裝置取得該車輛行駛在一私人道路上的車輛的至少一車輛週邊影像；以及一處理單元，耦接至該影像擷取裝置；其中，該處理單元被配置為執行以下步驟：對該車輛週邊影像進行一影像特徵萃取，以萃取出至少一關聯該私人道路上的至少一道路指示物件的特徵，該特徵至少包括該道路指示物件的一輪廓、一顏色、一文字、一符號、一圖像、一尺寸、一位置或前述的組合；對該特徵進行一座標轉換，以取得該道路指示物件相對於該車輛的一相對位置；根據該特徵以及該相對位置對該道路指示物件進行辨識，判斷是否符合一資料庫中的複數個預設條件的至少其中一者，以產生一辨識結果，其中，該資料庫的該預設條件是根據該私人道路的該道路指示物件而預先定義；以及響應該辨識結果在該車輛執行一操作。 A system for reminding road indication objects, comprising: at least one image capture device installed on a vehicle, the image capture device acquires at least one vehicle peripheral image of the vehicle driving on a private road; and a processing unit , coupled to the image capture device; Wherein, the processing unit is configured to perform the following steps: performing image feature extraction on the surrounding image of the vehicle to extract at least one feature associated with at least one road indicating object on the private road, the feature at least including the road indicating object An outline, a color, a character, a symbol, an image, a size, a position, or a combination of the above; carry out coordinate conversion on this feature, so as to obtain a relative position of the road indicating object relative to the vehicle; according to The feature and the relative position are used to identify the road indicating object, and determine whether at least one of a plurality of preset conditions in a database is met, so as to generate a recognition result, wherein the preset condition in the database is predefined according to the road indicating object of the private road; and performing an operation on the vehicle in response to the identification result.

如請求項5的系統，其中在執行該些步驟時，不使用一地圖資訊及/或該車輛的一定位資訊。 The system according to claim 5, wherein a map information and/or a location information of the vehicle are not used when performing the steps.

如請求項5的系統，其中該操作是經由一耦接至該處理單元的輸出單元對該車輛的一操作者發出一關聯於該辨識結果的提醒，該提醒為一視覺通知、一聲音通知、一觸覺回饋或前述的組合。 The system according to claim 5, wherein the operation is to issue a reminder related to the recognition result to an operator of the vehicle through an output unit coupled to the processing unit, the reminder is a visual notification, an audio notification, A haptic feedback or a combination of the foregoing.

如請求項5的系統，其中該操作是對該車輛進行一控制。The system of claim 5, wherein the operation is a control of the vehicle.