TWI619099B - Intelligent multifunctional driving assisted driving recording method and system - Google Patents

Abstract

The invention discloses a smart multi-functional driving assisted driving recording method and system, which uses an image capturing device to capture an image of driving road conditions during running of a vehicle. And the object feature data is established in the object feature database, and the object name is set in each object feature data. The information processing unit identifies the object name corresponding to the image of at least one of the objects of the driving road condition image in the object feature database, and then marks the object whose name has been identified in the driving road condition image displayed on the display screen. Marking as driving warning information to improve the safety of vehicles while driving.

Description

Intelligent multifunctional driving assisted driving recording method and system

The invention relates to a smart multi-functional driving assisted driving recording method and system, in particular to an image recognition technology capable of improving driving safety by driving warning.

In order to make the vehicle have the evidence-proven effect of traffic accidents, most of the owners will install a driving recorder in the vehicle. In addition to the driving road image in front of the vehicle, the corresponding time information can be recorded. Used as a basis for evidence of time backtracking. Although the conventional driving recorder is equipped with functions such as capturing traffic image and time information in front of the road; however, it cannot detect whether there is a pedestrian crossing the fast lane; or the movement of the preceding vehicle and the oncoming vehicle. Information, so that the relative motion relationship between the vehicle and the above objects cannot be determined. If a traffic accident occurs unfortunately, the traffic image recorded by the conventional driving recorder is actually less able to provide sufficient evidence to clarify the accident of the traffic accident. The reason, thus causing inconvenience and great troubles in the use of driving records.

In order to improve the above-mentioned deficiencies, the related art field has already obtained a patent as shown in the new Republic of China No. M414577 "Driving Navigation Recorder with Image Recognition Function"; it obtains the vehicle speed signal according to the latitude and longitude information of the vehicle and transmits it to the processing. The image capturing unit acquires the road image around the vehicle and transmits it to the processing unit; the image recognition unit analyzes the road image to generate an analysis result. The storage unit is installed in the driving navigation recorder host to store the vehicle speed signal, road image and analysis result; the image capturing unit captures the road image and transmits it everywhere The processing unit and the image recognition unit process the image recognition work, and the processing unit further connects the satellite positioning unit to obtain the current vehicle speed signal for the processing unit to recognize and determine whether it is a dangerous driving behavior.

Although the patent image recognition unit can analyze the road image to produce the analysis result; however, it does not provide any specific discussion on the specific technical content of the "analysis result" and whether it is one of the determining factors of "dangerous driving behavior"; In addition, the patent does not have the function of deep learning, so that it is impossible to enhance the classification function of the image comparison sample through deep learning. Therefore, the image image comparison accuracy is low and cannot be effectively applied.

There is also a patent as shown in the Republic of China Invention No. I495343 "Driving Recording System and Position Judgment Method"; it includes a curvature image lens, a computing module, a processing module display module, and a storage module. The curvature image captures the curvature image of the surrounding environment; the computation module restores the curvature image to the restored image. The coordinate positioning module receives the restored image, and restores the image by coordinates to generate a restored image by coordinate positioning; the processing module receives the restored image and generates time information; and the display module displays the restored image and time information. The storage module stores the restored image and time information.

Although the patent can calculate the relative motion relationship between the speed and the relative distance between the vehicle and at least one specific object from the restored image, as the basis for issuing the warning signal; however, it has no deep learning function, so that it cannot pass through. Deep learning to enhance the classification function of the image comparison sample, so the accuracy of the image image comparison identification; in addition, the patent only identifies the previous vehicle or the oncoming vehicle in motion to obtain the speed and Information such as distance; as far as pedestrians are concerned, it is impossible to monitor effectively, so it is impossible to detect whether there is a pedestrian who wants to cross the fast lane, so that it is impossible to effectively reduce the impact of driving pedestrians; or It is the information such as the speed and distance of the preceding vehicle and the oncoming vehicle during the movement, which causes great inconvenience in the use of the driving record.

In view of this, there is no patent for driving recording technology with object recognition for better deep learning calculus to enhance the classification function of the database, and the inventors have been continually based on the urgent needs of related industries. Under the circumstance of research and development, a set of inventions different from the above-mentioned prior art and patents has finally been developed.

A first object of the present invention is to provide a smart multi-functional driving assistance driving recording method and system, which can mainly monitor pedestrians in a vicinity of a lane or a moving vehicle in front of a moving vehicle and an oncoming vehicle, thereby monitoring results according to the monitoring result. Corresponding driving warnings can be made, and driving safety can be improved by driving warning. The technical means adopted for achieving the first object of the present invention is to capture an image of the driving road condition during running of the vehicle by using the image capturing device. The object feature data is created in the object feature database, and the object name is set in each object feature data. The information processing unit identifies the object name corresponding to the image of at least one of the objects of the driving road condition image in the object feature database, and then marks the object whose name has been identified in the driving road condition image displayed on the display screen. Mark as a driving warning information.

A second object of the present invention is to provide a smart multi-functional driving assisted driving recording method and system with deep learning function, which is mainly constructed by deep learning, in addition to strengthening the feature sample database classification function, and The feature extraction part of the previous paragraph uses the deep learning method to realize the self-learning correction, so that the high recognition rate can be achieved, and the traditional image recognition can avoid the loss of the recognition accuracy due to the environmental change of the previous segment. The technical means adopted for achieving the second object of the present invention is to use an image capturing device to capture an image of the driving road condition during running of the vehicle. The object feature data is created in the object feature database, and the object name is set in each object feature data. Information processing unit is identified in the object feature database The object name that matches the image of at least one of the objects of the road traffic image is recognized, and the object whose name has been identified is marked in the driving road condition image displayed on the display screen, and the marking can be used as the driving warning information. The method further includes a deep learning algorithm with a deep learning training function to perform the image recognition step, and the performing the deep learning algorithm includes the following steps: a training phase step, inputting distance data and a huge amount of the driving road condition The image is used to establish a deep learning model, and the depth learning model tests the correct rate of image recognition, and then determines whether the image recognition correct rate is sufficient. When the judgment result is yes, the identification result is output and stored; when the judgment result is no, And the depth learning model self-correction learning; and a running prediction phase step, the depth learning model inputs the distance data and the immediately captured driving road image, and the deep learning model performs predictive image recognition, Obtaining at least one object name and distance value of the identification result, and outputting the object name and the distance value.

10‧‧‧Image capture device

20‧‧‧ Object characterization database

21‧‧‧ Vehicle characteristics data

22‧‧‧Motorcycle/person profile data

23‧‧‧Pedestrian characteristics

24‧‧‧ Traffic Signature Characteristics

25‧‧‧ Coordinate Parameter Database

30‧‧‧Information Processing Unit

31‧‧‧Deep learning calculus module

310‧‧‧Deep learning model

32‧‧‧Danger Warning Evaluation Module

33‧‧‧ Driving hazard warning module

50‧‧‧Chassis

40‧‧‧ display screen

41‧‧‧ mark

51‧‧‧ memory device

52‧‧‧Audio playback module

ECU‧‧‧Driving computer

GPS‧‧‧Global Satellite Positioning Module

Ob‧‧‧ Object image

1 is a functional block diagram of a specific architecture of the present invention.

2 is a schematic diagram of the implementation of the training phase of the deep learning model of the present invention.

3 is a schematic diagram of the implementation of the operational prediction phase of the deep learning model of the present invention.

FIG. 4 is a schematic diagram of the implementation of the screen object selection in the present invention.

FIG. 5 is a schematic view showing the implementation of positioning the image of the object of the present invention into a known coordinate.

Fig. 6 is a schematic view showing another embodiment of the image of the object of the present invention nested in a known coordinate.

Fig. 7 is a schematic diagram showing the path prediction implementation of the image of the object of the present invention.

FIG. 8 is a schematic diagram of a specific implementation architecture of a convolutional neural network of the present invention.

Figure 9 is a schematic view showing the specific appearance of the present invention.

In order to allow the reviewing committee to further understand the technical features of the present invention and the technical means for achieving the object of the present invention, detailed description will be made with specific embodiments and drawings:

As shown in FIG. 1 and FIG. 4, in order to achieve the first object of the present invention, the present embodiment is an embodiment implemented by the image recognition method, which includes providing at least one image capturing device 10 and an object feature data. Technical features such as library 20, an information processing unit 30, and a display screen 40. The image capturing device 10 and the display screen 40 are both disposed in the vehicle for capturing and displaying the image of the driving road condition during the running of the vehicle. The object feature database 20 is configured to include a plurality of different object feature data, and set a feature data for each object feature data, and then set an object name for each feature data. When the information processing unit 30 executes, the following processing steps are sequentially included:

(a) Image recognition step: analyzing the feature value of the object image Ob of the driving road condition image after image pre-processing by image recognition, and then comparing the object feature database 20 with the feature value (such as the shape recognition) Characteristic data with a similarity of about 70% or more, and the corresponding object name is obtained from the feature data; if the image recognition success rate is not high, the similarity of the shape can be improved until the desired image recognition success rate is reached. .

(b) Position distance calculation step: Calculate the distance or relative position between the vehicle and the object (such as a pedestrian, a combination of a locomotive and a person or a vehicle) to output a corresponding distance signal or position signal.

(c) driving warning output step: as shown in FIG. 4, the object whose name has been identified is marked 41 in the driving road condition image displayed on the display screen 40, and the marking 41 is used as the driving warning information during driving; Specifically, the mark 41 is a sash for arranging objects, and different objects are selected by framing boxes of different shapes. Taking FIG. 4 as an example, a long rectangular frame represents a vehicle type, and a short rectangular frame represents a locomotive and The type of person combination, the oval frame represents the type of pedestrian; but not limited to this, the sash of different colors can also represent the object name. In addition, the present invention can output distance signals or position signals in an audio, voice, display or recording manner. Combined into the above-mentioned driving warning information, this driving warning information includes the object name information (such as vehicle, locomotive and person combination, pedestrian and traffic number), object location information, and object distance information, in addition to the above-mentioned mark 41. Traffic status, status information, speed information and hazard warning information.

According to the above, the traffic status real-time information is obtained through image recognition to know whether the traffic signal is in a green light or a red light. As far as the speed information is obtained, as shown in Figure 1, the position signal provided by the Global Positioning System (GPS) is calculated from the position change of a time unit; in addition, the signal acquisition method (such as can Bus) Take the vehicle's driving computer (ECU), and the vehicle speed information can also be obtained here, as shown in Figure 1.

Referring to the operation example shown in FIG. 4 to FIG. 6 , when the vehicle is running, the image capturing module 10 continuously captures at least two driving road image images, and pre-establishes a corresponding image of the traffic road image and The coordinate parameter database 25 containing the plurality of coordinate data, and then performing the image localization method to obtain the position of the center of gravity of the object image Ob, and then the coordinates of the coordinate parameter database 25 (ie, the coordinate frame shown in Figures 5-6) Each frame has its own defined coordinate value for comparison, and the coordinates of the object image Ob are calculated one by one, and the path information of the object image Ob is obtained as the object image Ob. Form a judgment basis for the intersection with the vehicle's driving route at a certain point. It can be seen from the embodiment shown in FIG. 5 that the present invention can obtain the position and the size of the object frame Ob by the object image Ob, and the volume size and distance of the object image Ob can be obtained (can be set by pre-determined parameters) Knowing; or using the two image capturing device 10 to perform the triangulation measurement method on the object to know the distance information between the object and the object; as for the time when the object speed information can be moved from the object image Ob to the next coordinate frame After the calculation, the speed information can be obtained. Then, the information processing unit 30 superimposes the text of the driving warning information into the road traffic image through the text superposition technology, except that the display screen 40 displays the superimposed driving warning information. In addition to the road traffic image of the text, the traffic device 51 can record the road traffic image superimposed with the driving warning information. It is explained that the technique of superimposing text on images is indeed a very well-known technique, so the text superposition technique is not described.

For example, Figures 4-6 show four objects, where object 1 (Ob1) is the same car in front, object 2 (Ob2) is the locomotive knight in front of the car, and object 3 (Ob3) is the right side of the lane. The pedestrian waiting for the crossing of the road, the object 4 (Ob4) is the opposite vehicle. When the image processing unit 30 recognizes and calculates the name, position and distance of each object, the information processing unit 30 passes through the display screen 40 or the audio playing module. 52 issued the road warning information, the driving warning information in addition to the markings 41 of the object in the driving road image display box, including the display or voice output "the front distance of 5 meters at the same place forward", "right front side There is a warning message such as the 2" organic car knight" and "Pedestrians at 25 meters from the right side of the slow lane, please drive carefully". Assuming that the distance between the vehicle and the preceding vehicle is less than 5 meters, as shown in FIG. 5, the information processing unit 30 transmits the information processing unit 30 through the display screen 40 or the voice hypothesis that the vehicle and the forward vehicle are less than 5 meters. Output "Please keep a safe distance from the preceding vehicle" through the display screen 40 or the audio playback module 52. Assuming that the pedestrian shuttles the road and is informed by the information processing unit 30 that the pedestrian will be hit, the information processing unit 30 outputs a danger warning of "Please slow down the speed, otherwise it will hit the pedestrian" through the display screen 40 or the audio playback module 52. News.

Referring to FIG. 1 to FIG. 4, in order to achieve the second embodiment of the second object of the present invention, at least one image capturing device 10, an object feature database 20, an information processing unit 30, and a display screen 40 are included. And other technical features. The image capturing device 10 is disposed in the vehicle for capturing an image of the driving road condition during running of the vehicle. The object feature database 20 is constructed to include a plurality of object feature data, and an object name is set for each object feature data. Specifically, the deep learning calculus module 31 can be a human intelligence algorithm such as a convolutional neural network (CNN) algorithm, an expert system algorithm, and a random forest algorithm. In addition, the deep learning calculus module 31 further includes the following processing steps when executed:

(a) Image recognition step: the object feature database 20 is compared with the image The name of the object to which at least one object image of the pre-processed road traffic image conforms. Specifically, the information processing unit 30 has a deep learning algorithm with a deep learning training function to perform an image recognition step, and sequentially performs feature extraction, feature selection, reasoning, and image processing on the road image processed by the image pre-processing. Predictive image recognition, and at least one feature captured image of the driving road condition image is processed into object feature data, and then the object feature data is accumulated into the object feature database 20 according to the category of the object for self-training learning. Thereby, at least one object name is obtained, and the feature value data is accumulated to enhance the classification function of the object feature database 20 and the function of enhancing the object feature extraction. Specifically, when the deep learning calculus module 31 is executed, the following phase steps are included:

(a-1) The training phase step, as shown in FIG. 2, establishes a deep learning model 310 for inputting distance data and a large amount of driving road condition image in the deep learning model 310, and testing the image recognition by the deep learning model 310. The correct rate is determined whether the image recognition correctness rate is sufficient. When the judgment result is yes, the identification result is output and stored; when the judgment result is negative, the deep learning model 310 is self-corrected and learned.

(a-2) running the prediction phase step, as shown in FIG. 3, inputting the distance data and the immediately captured driving road condition image in the deep learning model 310, and performing predictive image recognition by the deep learning model 310 to obtain at least one Identify the object name and distance value of the result, and then output the object name and distance value.

(b) Position distance calculation step: Calculate the distance or relative position between the vehicle and the object to output the corresponding distance signal or position signal.

(c) driving warning output step: marking the object whose name has been identified in the driving road condition image displayed on the display screen 40, and using the marking 41 as the driving warning information during driving; specifically, the above marking The 41 series is a sash for selecting objects, and the different objects are framed by sashes of different shapes; or by sashes of different colors, the driver can take the shape of the sash or Is the name of the object that the color recognizes; not only that, but also The sash of different shapes or different colors can be used as the display information of the driving warning level; or the object name can be represented by different audio; here, the driver can recognize the object relative warning from the shape or color of the sash. What is the rating?

The embodiment shown in FIG. 1 and FIG. 9 further includes a casing 50 for housing the image capturing device 10, the information processing unit 30, and the display screen 40. The image capturing device 10 is disposed in front of the casing 50. The back side is provided with a display screen 40, and the display screen 40 displays driving road condition images and driving warning information superimposed on the driving road condition image. In addition, the casing 50 is provided with a memory device 51 for recording driving road condition images and driving warning information. In addition, it should be noted that the image capturing device 10 is a complex lens image capturing device that can be used for ranging and recording. Here, the object and the vehicle can be obtained through the triangulation measurement method. The distance between them. Further, the triangular image localization measurement method may be a conventional Epipolar planar triangulation method, which mainly uses the multi-lens image capturing device 10 to capture the left driving road condition image and the right driving road condition image, and the information processing unit 10 Then the left and right road traffic images are calculated by the triangulation method of the Epipolar plane, and then the distance between the vehicle and the object Ob is obtained. Since the Epipolar plane triangulation method is a very well-known technique, it is no longer The specific operation content will be described.

Referring to the embodiment shown in FIG. 1 and FIG. 7, the information processing unit 30 includes a hazard warning evaluation module 32, and the hazard warning information is the evaluation calculated by the hazard warning evaluation module 32, which is the predicted path D of the object. ₁ Whether or not a crossing point is formed at a time point with the vehicle's driving route D ₂ , and if the judgment result is yes, a danger warning message is issued. Furthermore, the object feature database 20 shown in FIG. 1 includes a vehicle feature data 21 storing a plurality of vehicle feature samples, a locomotive/person profile data storing a plurality of locomotive and human combination feature samples, and a stored plurality of pedestrian feature samples. The pedestrian characteristic data 23 and a traffic sign feature data 24 for storing a plurality of traffic sign feature samples.

Referring to FIG. 1 , the information processing unit 30 includes a row of vehicle danger level warning module 33 , and the driving danger level warning module 33 determines whether the moving path of the object is related to the vehicle. The driving line analyzes the possibility of forming a meeting point at a certain point in time, and outputs the driving warning information according to the ranking of the hazard levels evaluated by the objects. Take 4 and 7 as an example to illustrate, the locomotive in Figure 4 is listed as the highest level of driving warning information, such as voice and recording output: "Be careful of the driving line of the front locomotive knight"; secondly, in Figure 4 The same-direction car is listed as the second-level driving warning information, such as voice and recording output: "Be careful to maintain the safe distance from the vehicle in front"; if the pedestrian in Figure 4 stops waiting, the pedestrian is listed as the lowest. The level of driving warning information, such as voice and record output: "Please continue to pay attention to the right side of the pedestrian line."

Through the description of the above specific embodiments, it is known that the second embodiment of the present invention uses a deep learning technique as a base for the front identification of an auxiliary system for advanced driving, thereby identifying cars, pedestrians, locomotives, and the like in driving road conditions images. Traffic signs and other objects, in order to achieve the effect of driving warning, and the deep learning algorithm is built in the mobile device-specific chipset (including graphics processor GPU), in order to achieve the purpose of real-time detection of driving dynamics, and finally packaged into advanced The driver identification system's previous image recognition solution is provided to the relevant industry. Furthermore, the Convolutional Neural Network (CNN) in deep learning has broken through the recognition rates of early classifiers and methods after 2012. In general, traditional image recognition often loses the recognition accuracy because the feature extraction algorithm in the previous segment is susceptible to environmental changes. To this end, the present invention is to improve this deficiency, using the Convolutional Neural Network (CNN) to use the deep learning method from the learning section of the previous paragraph to learn the correction, as long as a sufficient number of objects are established. The feature data allows the deep learning model to self-learn and correct. In this way, it can achieve a high recognition rate and can break through the traditional traditional machine learning methods.

Generally, the convolutional neural network obtains the driving road condition image from the image capturing device 10, and then undergoes image pre-processing (ie, pre-processing), feature extraction, feature selection, and then reasoning and predictive identification. On the other hand, the deep learning essence of convolutional neural networks is to build more useful features by constructing machine learning models with multiple hidden layers and massive training data, so as to improve the accuracy of classification or prediction. Convolutional neural networks use massive training numbers According to the study of feature identification, it is possible to portray the rich internal information of the data. Since the convolutional neural network is a weight-sharing network structure, in addition to reducing the complexity of the network model, the number of weights can be reduced. This advantage is more obvious when the input of the network is a multi-dimensional image, so that the image can be directly used as the input of the network, avoiding the complicated feature extraction and data reconstruction process in the traditional image recognition algorithm. Object classification is almost always based on statistical features, which means that certain features must be extracted before resolving. However, explicit feature extraction is not easy and is not always reliable in some application problems. The convolutional neural network avoids explicit feature sampling and implicitly learns from the training data. This makes the convolutional neural network distinct from other neural network-based classifiers, merging feature extraction functions into multilayer perceptrons through structural reorganization and reduced weights. It can process grayscale images directly and can be used directly to process image-based classifications. The convolutional network has the following advantages in image processing compared with the general neural network: the input image and the network topology can be well matched; the feature extraction is performed simultaneously with the pattern classification, and simultaneously generated in the training; Sharing can reduce the training parameters of the network, making the neural network structure simpler and more adaptable.

Please refer to FIG. 8 , which is a schematic diagram of a specific model structure of the convolutional neural network. The input driving road condition image is convoluted by the three trainable filters and the addable bias, and the convolution processing is performed on the C1 layer. Three feature maps are generated, and then four pixels of each group in the feature map are summed, weighted, and offset; then, a feature map of three S2 layers is obtained through a function (Sigmoid). The map is filtered again to obtain the C3 layer. This hierarchical structure produces S4...Sn as well as S2. Finally, these pixel values are rasterized and concatenated into a vector input to a traditional neural network to obtain an output.

In general, the C layer is a feature extraction layer, and the input of each neuron is connected to the local receptive field of the previous layer, and the local feature is extracted, and once the local feature is captured, it is interposed with other features. The positional relationship is also determined. As for the S layer, it is a feature mapping layer. Each computing layer of the network is composed of multiple feature maps, each of which is mapped to a plane, on a plane. All neurons have equal weights. The feature mapping structure uses a small sigmoid function that affects the function kernel as the activation function of the convolution network (Activation function; (1) Tan H, (2) Rectifier Relu), (3) Parameteric Rect, fied Lineor Unit (Prelu), The feature map has displacement invariance. In addition, since the neurons on a mapped surface share weights, the number of network free parameters is reduced, which reduces the complexity of network parameter selection. Each feature extraction layer (C-layer) in the convolutional neural network is followed by a computational layer (S-layer) for local averaging and secondary extraction. This unique feature extraction is performed twice. The structure allows the network to have high distortion tolerance for input samples when it is identified.

Convolutional neural networks are primarily used to identify two-dimensional graphics of displacement, scaling, and other forms of distortion invariance. Since the feature detection layer of the convolutional neural network learns through the training data, when using the convolutional neural network, explicit feature extraction is avoided, and learning is implicitly learned from the training data; The weights of the neurons on the feature mapping surface are the same, so the network can learn in parallel, which is also a major advantage of the convolutional neural network relative to the neural network connected to each other. The convolutional neural network has unique advantages in speech recognition and image processing with its special structure of local weight sharing. Its layout is closer to the actual biological neural network, and weight sharing reduces the complexity of the network. In particular, the image of the multi-dimensional input vector can be directly input into the network, which avoids the complexity of data reconstruction in the feature extraction and classification process.

The above is only a possible embodiment of the present invention, and is not intended to limit the scope of the patents of the present invention, and the equivalent implementations of other changes according to the contents, features and spirits of the following claims should be It is included in the patent of the present invention. The invention is specifically defined in the structural features of the request item, is not found in the same kind of articles, and has practicality and progress, has met the requirements of the invention patent, and has filed an application according to law, and invites the bureau to approve the patent according to law to maintain the present invention. The legal rights of the applicant.

Claims (9)

A smart multi-functional driving assistance driving recording method includes: providing at least one image capturing device, an object feature database, an information processing unit, and a display screen; the image capturing device is disposed on a vehicle for Taking the driving road condition image of the vehicle while driving; the display screen is installed in the vehicle to display the driving road condition image; and the object feature database is established with a plurality of different object feature data, and each of the The object feature data is set with an object name; and the information processing unit includes the following processing steps in sequence: (a) image recognition step: identifying the feature of the object feature database and the image of the traffic road image The object name of the at least one object conforms; and (b) the driving warning output step: marking the object that has identified the object name in the driving road condition image displayed by the display screen, and marking the object as the marking Driving warning information while driving; wherein the information processing unit further includes a deep learning training function to perform the image recognition a deep learning calculus module for sequentially performing feature extraction, feature selection, reasoning, and predictive image recognition on the driving road condition image processed by the image, and at least one of the features of the driving road condition image Part of the image processing is taken as the feature data of the object, and the feature data of the object is accumulated in the object feature database according to the category of the object, so as to be trained by the deep learning calculus module to obtain at least one object name. The method of claim 1, wherein the marking is selected from the group consisting of a sash for framing the object and an audio, the object being selected from the sash of different shapes; or in different colors The sash box selects the object; or the different audio to represent the object name. The method of claim 1, wherein the performing the deep learning calculus module comprises the following steps: In a training phase step, a deep learning model is established, and the distance learning model inputs distance data and a huge amount of the driving road condition image, and the depth learning model tests the correct rate of image recognition, and then determines whether the image recognition correct rate is correct. Sufficiently, when the judgment result is yes, the identification result is output and stored; when the judgment result is no, the deep learning model is self-corrected learning; and a running prediction phase step, the deep learning model inputs the deep learning model The distance information and the immediately captured image of the driving road condition, and the predictive image recognition is performed by the deep learning model to obtain the object name and the distance value of the at least one identification result, and then outputting the object name and the distance value. The method of claim 1, further comprising a position distance calculation step of calculating a distance or a relative position between the vehicle and the object, and outputting the display screen or an audio playback module The distance between the vehicle and the object or the audio signal of the location. The method of claim 1, wherein the object feature database includes vehicle feature data for storing a plurality of vehicle feature samples, a locomotive/person profile data for storing a plurality of locomotive and human combination feature samples, and a stored plurality of pedestrian feature samples. Pedestrian characteristics and a traffic log characteristic data for storing a plurality of traffic log feature samples. The method of claim 1, further comprising a casing for housing the image capturing device, the object feature database, and the information processing unit; the image capturing device is disposed on the front surface of the casing a display screen is displayed, the display screen displays the driving road condition image and the driving warning information superimposed on the driving road condition image; and the storage device is provided with a memory device for recording the driving road condition image and the driving warning information The image capturing device is a complex lens image capturing device for ranging and recording. The method of claim 1, wherein the information processing unit comprises a line of vehicle hazard warning module, and the driving danger level warning module forms a meeting with the moving path of the object at a time point of the driving line of the vehicle. The probability of the point is predicted and evaluated, and the driving warning information is output according to the ranking of the hazard levels evaluated by the object. A smart multi-functional driving assistance driving record system comprising: at least one image capturing device disposed on a vehicle for capturing an image of driving road conditions while the vehicle is traveling; and a display screen disposed in the vehicle The utility model is configured to display the driving road condition image; the object feature database is configured to include a plurality of object feature data, and an object name is set in each of the object feature data; and an information processing unit is executed in sequence The method includes the following steps: (a) an image recognition step: comparing the object name with the feature of the object feature database to match the feature of at least one object in the driving road image; and (b) driving warning output Step: marking the object that has identified the object name in the driving road condition image displayed by the display screen, and using the marking as the driving warning information during driving; wherein the information processing unit further includes a depth Learning a training function to perform a deep learning algorithm module of the image recognition step, in accordance with the driving road condition image processed by the image Performing feature extraction, feature selection, reasoning, and predictive image recognition, and processing at least one of the feature image captured by the driving road condition image as the feature data of the object, and then selecting the object feature data according to the category of the object. Accumulating in the object feature database for the deep learning calculus module training learning, thereby obtaining at least one object name. The system of claim 8, wherein the mark is a frame for arranging the object, and the object is selected by a frame of different shapes or by a frame of a different color. object.