CN110909674A - Traffic sign identification method, device, equipment and storage medium - Google Patents
Traffic sign identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN110909674A CN110909674A CN201911149014.9A CN201911149014A CN110909674A CN 110909674 A CN110909674 A CN 110909674A CN 201911149014 A CN201911149014 A CN 201911149014A CN 110909674 A CN110909674 A CN 110909674A
- Authority
- CN
- China
- Prior art keywords
- identification
- traffic sign
- picture
- target
- recognition
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000012216 screening Methods 0.000 claims abstract description 16
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013434 data augmentation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a traffic sign identification method, a device, equipment and a storage medium. The method comprises the following steps: acquiring pictures in an identification picture group to identify large classes of traffic signs, and acquiring a target picture of which an identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the large classes of the traffic signs identified in the identification region and an identification confidence coefficient; screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture; and performing traffic sign fine classification recognition in the at least one target judgment area, and acquiring traffic sign fine classification which meets a second confidence degree condition as a traffic sign recognition result. By using the technical scheme of the embodiment of the invention, the large categories, the numerical values and other fine categories of the traffic signs can be identified in real time, and the early warning is timely carried out on the driver, so that the driving safety is improved.
Description
Technical Field
The embodiment of the invention relates to an image identification technology, in particular to a traffic sign identification method, a device, equipment and a storage medium.
Background
When conditions such as traffic control, road construction modification and complex road conditions occur, drivers are easy to miss in forbidden areas and overspeed driving due to negligence or misjudgment of traffic signs, and even serious consequences of traffic accidents are caused.
In the prior art, various identification models are generally adopted to identify traffic signs and give early warning to drivers. An existing traffic sign recognition model, for example, an SSD (Single Shot multi box Detector), shows a practical operation result, and although the type of the traffic sign can be normally recognized, a specific numerical value related to the traffic sign, such as a numerical value recognition error rate of a speed limit, is high. At this time, if the recognition of the traffic sign recognition model is wrong, the driver is easy to misjudge, so that the problems of overspeed, overweight driving and the like are caused.
Disclosure of Invention
The embodiment of the invention provides a traffic sign identification method, a device, equipment and a storage medium, which are used for realizing real-time, real and accurate identification of a traffic sign and timely early warning for a driver in the driving process of the driver.
In a first aspect, an embodiment of the present invention provides a traffic sign identification method, where the method includes:
acquiring pictures in an identification picture group to identify large classes of traffic signs, and acquiring a target picture of which an identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the large classes of the traffic signs identified in the identification region and an identification confidence coefficient;
screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture;
and performing traffic sign fine classification recognition in the at least one target judgment area, and acquiring traffic sign fine classification which meets a second confidence degree condition as a traffic sign recognition result.
Further, acquiring pictures in the recognition picture group for traffic sign large category recognition, and acquiring a target picture of which the recognition result meets a first confidence degree condition comprises:
acquiring a picture in the middle position from the identification picture group as a basic picture;
carrying out traffic sign large category identification on the basic picture, and acquiring the traffic sign large category and the identification confidence coefficient identified in each identification area;
and if the recognition confidence of the basic picture exceeds a preset first confidence threshold, respectively performing traffic sign large category recognition on the rest pictures in the recognition picture group, and acquiring the picture with the highest recognition confidence as the target picture.
Further, in each identification area of the target picture, at least one target judgment area meeting a traffic sign identification condition is screened out, and the at least one target judgment area comprises at least one of the following items:
if the length-width ratio of the traffic sign contained in the target identification area is determined to be abnormal, deleting the target identification area;
if the sum of the lengths and the widths of the traffic signs contained in the target identification area is smaller than the sum of the preset lengths and widths, deleting the target identification area; and
and if the traffic sign in the target identification area in which the forbidden traffic sign is identified is determined not to meet the forbidden traffic sign condition, deleting the target identification area.
Further, determining that the traffic sign in the target identification area in which the forbidden traffic sign is identified does not satisfy the forbidden traffic sign condition includes:
converting the target identification area into a hexagonal pyramid model format;
counting the number of red pixel points in the target identification area according to the hue parameters of the pixel points;
and if the number of the red pixel points is less than the preset number of pixels, determining that the traffic sign contained in the target identification area does not meet the forbidden traffic sign condition.
Further, performing fine traffic sign category identification in the at least one target judgment area, and acquiring a fine traffic sign category meeting a second confidence condition as a traffic sign identification result, including:
performing traffic sign fine classification identification on the at least one target judgment area to obtain an identification result, wherein the identification result comprises the traffic sign fine classification identified in the target judgment area and an identification confidence coefficient;
and if the recognition confidence of the target judgment area exceeds a preset second confidence threshold, the recognized traffic sign fine classification meets a second confidence condition, and the recognized traffic sign fine classification is used as a traffic sign recognition result.
Further, before acquiring the pictures in the recognition picture group for traffic sign large category recognition, the method further comprises:
cutting a road picture shot in real time, and storing the cut road picture into a picture queue;
acquiring a set number of road pictures in the picture queue as the identification picture group;
after acquiring the recognition result meeting the second confidence condition as the traffic sign recognition result, the method further comprises the following steps:
and returning to execute the operation of acquiring the road pictures with the set number in the picture queue as the identification picture group until the identification finishing condition is met.
Further, the cutting of the road picture shot in real time includes:
and cutting the shot road picture into a road picture with a preset size, wherein the cut road picture is the upper right part of the shot road picture.
In a second aspect, an embodiment of the present invention further provides a traffic sign identification apparatus, where the apparatus includes:
the traffic sign large category identification module is used for acquiring pictures in the identification picture group to identify the traffic sign large category and acquiring a target picture of which the identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the traffic sign large category identified in the identification region and an identification confidence coefficient;
the identification condition screening module is used for screening out at least one target judgment area meeting the identification condition of the traffic sign in each identification area of the target picture;
and the traffic sign fine classification identification module is used for carrying out traffic sign fine classification identification in the at least one target judgment area and acquiring the traffic sign fine classification meeting a second confidence coefficient condition as a traffic sign identification result.
In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the traffic sign identification method according to any embodiment of the present invention.
In a fourth aspect, embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the program, when executed by a computer processor, implements a traffic sign recognition method as provided by any of the embodiments of the present invention.
The embodiment of the invention obtains the identification area by carrying out traffic sign large-class identification on the target picture, screens the identification area, and then carries out traffic sign fine-class identification on the screened target judgment area, thereby obtaining the traffic sign fine-class classification. The problem of low accuracy in identifying fine traffic sign categories after identifying the large traffic sign categories in the prior art is solved, and the effect of accurately identifying the large traffic sign categories and the fine traffic sign categories in real time is achieved.
Drawings
FIG. 1 is a flow chart of a traffic sign recognition method according to a first embodiment of the present invention;
FIG. 2 is a flow chart of a traffic sign recognition method according to a second embodiment of the present invention;
FIG. 3 is a flow chart of traffic sign recognition suitable for use in embodiments of the present invention;
fig. 4 is a schematic structural diagram of a traffic sign recognition apparatus according to a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of a traffic sign recognition apparatus in a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a traffic sign recognition method according to an embodiment of the present invention, where the embodiment is applicable to a situation where a driver recognizes a traffic sign in real time and warns the driver of the traffic sign in driving, and the method may be executed by a traffic sign recognition apparatus, which may be implemented by software and/or hardware, and may be generally integrated in a vehicle. The method specifically comprises the following steps:
step 110, acquiring pictures in the recognition picture group to perform traffic sign large category recognition, and acquiring a target picture of which the recognition result meets a first confidence coefficient condition;
the identification result comprises at least one identification area, a large category of the traffic sign identified in the identification area and an identification confidence coefficient;
the large categories of the traffic signs mainly comprise speed limit, driving limit, height limit, weight limit, speed measurement and the like, the categories of the traffic signs are accurately identified and early-warned in time, and the driving safety of a driver can be guaranteed. Confidence, also referred to as reliability, confidence level, or confidence coefficient, etc., refers to the probability that an overall parameter value falls within a certain region of the sample statistics. The recognition result of the target picture meets the confidence condition, which shows that the traffic sign large category recognized in the recognition area of the target picture is basically reliable. The identification area is a part which is intercepted and possibly contains a traffic sign in the process of identifying the target picture. In a specific embodiment, the pixel size of the target picture may be 300 × 300, and the size of the identification area may be defined as 32 × 32 in this case, but the size of the identification area is not limited in this embodiment.
Wherein the traffic sign large category identification is accomplished through a deep neural network model one. The skeleton network structure of the first model can be any one of or a variety of convolutional neural networks. For example: VGG (Visual geometry group) model, ResNet (Deep residual Network), densnet (Dense Convolutional Network), and the like, and the framework Network structure of the first model is not limited in this embodiment. In one specific embodiment, the model-skeleton network uses MobileNet V2, and the training parameters are set as: the size of the picture input pixel is 300 × 300, the training times are 300000, random cutting and left-right turning are adopted for data augmentation, the length-width ratio of an anchor frame is set to be 1:1, 2:1, 1:2, 1:3 and 3:1, the initial value of the learning rate is 0.004, and after 1000 times of training, the learning rate is adjusted to be 0.95. And after 200000 iterations, the graph rewriter is used for quantization, the number of active bits is set to 8, the number of weight bits is set to 8, and the quantization can compress the model so as to occupy less memory and have high running speed. And the trained first model is responsible for carrying out traffic sign large-class identification on the pictures in the identification picture group, and when the pictures are input, the output of the first model is the large classes of traffic signs possibly contained in each identification area in the pictures and the corresponding confidence coefficients of the traffic signs. Step 120, screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture;
if the length-width ratio of the traffic sign contained in the target identification area is determined to be abnormal, deleting the target identification area; the road traffic sign has the advantages that in real life, the aspect ratio of the road traffic sign is 1:1, so that when the aspect ratio of the traffic sign in the target recognition area is too large or too small, the error recognition can be basically judged, and the traffic sign is removed from the recognition area, so that the workload of later-stage traffic sign fine classification recognition can be reduced. In a specific embodiment, the aspect ratio abnormality may be specified as an aspect ratio abnormality when the aspect ratio is less than 0.7 or the aspect ratio is greater than 1.5, but the specific threshold value of the aspect ratio abnormality is not restrictively specified in the present embodiment.
If the sum of the lengths and the widths of the traffic signs contained in the target identification area is smaller than the sum of the preset lengths and widths, deleting the target identification area; the arrangement has the advantages that when the sum of the length and the width of the traffic sign is smaller, the proportion of the pixel number of the traffic sign to the pixel number of the whole recognition area is lower, the large category and the fine category of the traffic sign are easily judged by mistake, and the traffic sign recognition accuracy can be ensured by removing the traffic sign from the recognition area. In a specific embodiment, the size of the identification area may be defined as 32 × 32, and if the sum of the lengths and widths of the identified traffic signs is less than 28, it indicates that the proportion of the traffic signs in the identification area is too small and needs to be removed.
And if the traffic sign in the target identification area in which the forbidden traffic sign is identified is determined not to meet the forbidden traffic sign condition, deleting the target identification area. The method has the advantages that the ground color of forbidden traffic signs such as speed limit, weight limit, height limit, restriction and the like is red, if forbidden traffic signs are identified in the identification area, the number of red pixel points in the identification area is too small, so that misjudgment is possible at the moment, and the traffic signs are removed from the identification area, so that the accuracy of identification of the traffic signs is improved.
The method for determining that the traffic sign in the target identification area for identifying the forbidden traffic sign does not meet the forbidden traffic sign condition comprises the following steps: converting the target identification area into a hexagonal pyramid model format; counting the number of red pixel points in the target identification area according to the hue parameters of the pixel points; and if the number of the red pixel points is less than the preset number of pixels, determining that the traffic sign contained in the target identification area does not meet the forbidden traffic sign condition.
The hexagonal pyramid model format is also called as HSV (Hue, Saturation, brightness) color model, and the HSV color model expresses a color image by three parameters of Hue, Saturation and brightness, can intuitively express Hue, brightness and brightness of a color, and is convenient for color contrast. The target identification area is originally in a space domain for expressing the color image by an RGB (Red, Green, Blue, Red, Green and Blue) model, and is converted into the space domain for expressing the color image by an HSV (hue, saturation, value) model, so that the number of Red pixel points can be counted conveniently. In the HSV model, the value of the H parameter is 0-360, wherein pixel points with the H value smaller than 15 and the H value larger than 345 are red pixel points. And counting the number of the pixels with the H value in the range as the number of the red pixels. In a specific embodiment, when the identification area is set to 32 × 32 and the number of red pixels is less than 80, it may be determined that a false determination occurs, and the false determination is removed from the identification area.
And step 130, performing traffic sign fine classification recognition in the at least one target judgment area, and acquiring traffic sign fine classification satisfying a second confidence degree condition as a traffic sign recognition result.
Wherein, the fine classification of the traffic signs is identified for the target judgment area screened in the step 120. The traffic sign fine category, i.e., the traffic sign large category and the value or letter following the traffic sign large category, is, for example, the speed limit 30 when the traffic sign large category is the speed limit. If the identification of the traffic sign fine category meets the second confidence condition, the traffic sign fine category identified in the step is basically reliable, and can be used as a traffic sign identification result to prompt a driver.
The traffic sign fine classification identification is completed through a classification neural network model II. In a specific embodiment, the model two network adopts DenseNet, so that the network has the advantages of adopting a characteristic diagram to carry out connection across network layers, less model parameters and high accuracy. The model training parameters are set as: the depth of the network is set to 40, the dimension of each layer is reduced by convolution with 1X1 before input, and the dimension of each layer is reduced by convolution with 1X1 between different DenseNet modules. The learning rate is set to 0.0001 and the translation layer compression factor is set to 0.5. And the trained model II is responsible for identifying the traffic sign fine classification, and when the traffic sign fine classification is input into a target judgment area containing the traffic sign, the traffic sign fine classification of each target judgment area and the corresponding confidence coefficient of the traffic sign fine classification are output.
According to the technical scheme of the embodiment, the traffic sign large-class identification is carried out on the target picture to obtain the identification area, the identification area is screened, and then the traffic sign fine-class identification is carried out on the screened target judgment area to obtain the traffic sign fine-class. The problem of low accuracy in identifying fine traffic sign categories after identifying the large traffic sign categories in the prior art is solved, and the effect of accurately identifying the large traffic sign categories and the fine traffic sign categories in real time is achieved.
Example two
Fig. 2 is a flowchart of a traffic sign identification method in the second embodiment of the present invention, and this embodiment further details a traffic sign large category identification process and a traffic sign fine category identification process on the basis of the technical solution in the second embodiment, and specifically includes the following steps:
step 210, cutting the road picture shot in real time, and storing the cut road picture into a picture queue;
the method comprises the steps of obtaining a road picture, wherein the shot road picture is cut into a road picture with a preset size, and the cut road picture is the upper right part of the shot road picture.
The road traffic sign board is arranged on the right side of the road in general, and the vehicle runs on the right side, so that the probability that the traffic sign appears on the lower left side of the road picture is extremely low, the shot road picture is cut in advance, the upper right part of the road picture is reserved, and the workload of the subsequent identification process can be reduced. In a specific embodiment, the picture size may be cut to 300 × 300, but the present embodiment does not limit the picture size.
Step 220, acquiring a set number of road pictures in the picture queue as the identification picture group;
the road pictures with the set number are placed into the identification picture group for processing, and the arrangement has the advantages that the road pictures are shot in real time in the driving process of the vehicle, so that the number of the road pictures is huge, and the workload can be reduced by processing the road pictures with the set number as one group. Meanwhile, because the contents of the continuously shot photos are generally continuous, the influence of the grouping processing on the identification accuracy rate can be ignored.
Fig. 3 is a flow chart of traffic sign recognition applicable to the embodiment of the present invention, and as shown in fig. 3, an odd number of pictures are taken as a recognition picture group. However, the present embodiment is not limited to whether the set number is an odd number or an even number.
Step 230, obtaining a picture in the middle position in the identification picture group as a basic picture;
the method has the advantages that the pictures at the middle positions can represent the identification picture group most because the contents of the continuously shot road pictures are continuous generally, and if the traffic signs are not identified in the pictures at the middle positions, the traffic signs do not appear in other pictures in the identification picture group.
Step 240, performing traffic sign large category identification on the basic picture, and acquiring the traffic sign large categories and identification confidence degrees identified in each identification area;
when the pictures at the middle position, that is, the basic pictures are subjected to traffic sign large category identification, one basic picture may include a plurality of areas where traffic signs may appear, and the areas where the traffic signs may appear are cut to serve as identification areas. And respectively carrying out traffic sign large-class identification on the identification areas and calculating confidence degrees.
Step 250, if the recognition confidence of the basic picture exceeds a preset first confidence threshold, respectively performing traffic sign large category recognition on the rest pictures in the recognition picture group, and acquiring a picture with the highest recognition confidence as the target picture;
the identification confidence of the basic picture exceeds a preset first confidence threshold, which indicates that the reliability of the traffic sign appearing in the basic picture is higher, so that the traffic sign large category identification is also carried out on other pictures in the picture identification group, the confidence is calculated, and the picture with the highest confidence in the picture identification group is obtained as the target picture.
As shown in fig. 3, the pictures at the middle position are taken, the calculation of the large category and the confidence coefficient of the traffic sign is carried out on the pictures through the model, if the confidence coefficient is less than 0.6, the calculation is directly returned, and an odd number of pictures are taken as an identification picture group in the picture set. If the confidence coefficient is larger than 0.6, the next operation can be carried out, namely, the traffic sign large-class identification and confidence coefficient calculation are carried out on the rest pictures in the identification picture group through the model, the picture with the highest confidence coefficient is selected as the target picture, and at least one identification region of the target picture is obtained.
Step 260, screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture;
step 270, performing traffic sign fine classification identification on the at least one target judgment area to obtain an identification result, wherein the identification result comprises the traffic sign fine classification identified in the target judgment area and an identification confidence;
and carrying out traffic sign fine classification identification on the screened target judgment area, identifying traffic sign fine classification in the target judgment area and calculating confidence.
Step 280, if the recognition confidence of the target judgment region exceeds a preset second confidence threshold, the recognized traffic sign fine classification meets a second confidence condition, and the recognized traffic sign fine classification is used as a traffic sign recognition result.
If the identification of the traffic sign fine category meets the second confidence condition, the traffic sign fine category identified in the step is basically reliable, and can be used as a traffic sign identification result to prompt a driver.
As shown in fig. 3, after the two screened target judgment areas of the model are subjected to traffic sign fine classification recognition and confidence calculation, if the confidence is greater than 0.96, the traffic sign fine classification is output as a traffic sign recognition result.
If the identification process is not finished, the set of identification picture groups are identified and the traffic sign identification result is obtained, then the set number of road pictures are obtained again to serve as a new identification picture group, and the identification process is continued.
Step 2100, end recognition process.
According to the technical scheme, the pictures are subjected to cutting preprocessing, a set number of preprocessed pictures are taken as a picture group, whether the pictures at the middle position of the picture group can identify the large class of the traffic sign is judged, the confidence coefficient is calculated, if the confidence coefficient meets the requirement, the rest pictures in the picture group are identified and the confidence coefficient is calculated, the picture with the highest confidence coefficient is selected, the identification region is screened, the screened target judgment region is subjected to identification of the fine class of the traffic sign and calculation of the confidence coefficient again, and if the confidence coefficient of the target judgment region meets the requirement, the identified fine class of the traffic sign is taken as a final result. The problem of among the prior art, although can discern the major classification of traffic sign, the error rate is higher when discerning the thin classification of traffic sign. The effect of accurately identifying the large categories and the fine categories of the traffic signs is achieved.
EXAMPLE III
Fig. 4 is a schematic structural diagram of a traffic sign recognition apparatus according to a third embodiment of the present invention, where the traffic sign recognition apparatus includes: a traffic sign large category identification module 310, an identification condition screening module 320 and a traffic sign fine category identification module 330, wherein:
the traffic sign large category identification module 310 is configured to acquire pictures in the identification picture group to perform traffic sign large category identification, and acquire a target picture whose identification result meets a first confidence condition, where the identification result includes at least one identification region, a traffic sign large category identified in the identification region, and an identification confidence;
the identification condition screening module 320 is configured to screen out at least one target judgment region that meets a traffic sign identification condition in each identification region of the target picture;
and the traffic sign fine classification identification module 330 is configured to perform traffic sign fine classification identification in the at least one target judgment area, and acquire a traffic sign fine classification meeting a second confidence condition as a traffic sign identification result.
On the basis of the above embodiment, the traffic sign large category identification module 310 includes:
a basic picture acquiring unit, configured to acquire, as a basic picture, a picture at an intermediate position in the identification picture group;
the basic picture large category identification unit is used for carrying out traffic sign large category identification on the basic picture, and obtaining the traffic sign large category and the identification confidence coefficient identified in each identification area;
and the other picture large category identification unit is used for respectively carrying out traffic sign large category identification on the other pictures in the identification picture group and acquiring the picture with the highest identification confidence coefficient as the target picture if the identification confidence coefficient of the basic picture exceeds a preset first confidence coefficient threshold value.
On the basis of the above embodiment, the condition screening module 320 is identified, which includes:
the first target identification area deleting unit is used for deleting the target identification area if the traffic sign length-width ratio contained in the target identification area is determined to be abnormal;
the second target identification area deleting unit is used for deleting the target identification area if the sum of the lengths and the widths of the traffic signs contained in the target identification area is smaller than the sum of the preset lengths and the widths;
and the third target identification area deleting unit is used for deleting the target identification area if the traffic sign in the target identification area in which the forbidden traffic sign is identified is determined not to meet the forbidden traffic sign condition.
On the basis of the above embodiment, the third target recognition area deleting unit includes:
the format conversion subunit is used for converting the target identification area into a hexagonal pyramid model format;
the pixel point number counting subunit is used for counting the number of the red pixel points in the target identification area according to the hue parameters of the pixel points;
and the forbidden traffic sign condition judgment subunit is used for determining that the traffic signs contained in the target identification area do not meet the forbidden traffic sign condition if the number of the red pixel points is less than the preset pixel number.
On the basis of the above embodiment, the traffic sign fine category identification module 330 includes:
the traffic sign fine classification identification unit is used for carrying out traffic sign fine classification identification on the at least one target judgment area to obtain an identification result, and the identification result comprises the traffic sign fine classification identified in the target judgment area and an identification confidence coefficient;
and the second confidence condition judging unit is used for judging whether the identification confidence of the target judging area exceeds a preset second confidence threshold value or not, if so, judging that the identified traffic sign fine classification meets the second confidence condition, and taking the identified traffic sign fine classification as a traffic sign identification result.
On the basis of the above embodiment, the traffic sign recognition apparatus further includes:
the picture cutting module is used for cutting the road picture shot in real time and storing the cut road picture into a picture queue;
the identification picture group acquisition module is used for acquiring road pictures with set quantity in the picture queue as the identification picture group;
and the return execution module is used for returning and executing the operation of acquiring the set number of road pictures in the picture queue as the identification picture group until the identification finishing condition is met.
On the basis of the above embodiment, the picture cropping module includes:
and the road photo cutting unit is used for cutting the shot road photo into a road photo with a preset size, wherein the cut road photo is the upper right part of the shot road photo.
The traffic sign recognition device provided by the embodiment of the invention can execute the traffic sign recognition method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 5 is a schematic structural diagram of an apparatus according to a fourth embodiment of the present invention, as shown in fig. 5, the apparatus includes a processor 40, a memory 41, an input device 42, and an output device 43; the number of processors 40 in the device may be one or more, and one processor 40 is taken as an example in fig. 5; the processor 40, the memory 41, the input device 42 and the output device 43 in the apparatus may be connected by a bus or other means, which is exemplified in fig. 5.
The memory 41, as a computer-readable storage medium, may be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the traffic sign recognition method in the embodiment of the present invention (for example, the traffic sign large category recognition module 310, the recognition condition screening module 320, and the traffic sign small category recognition module 330 in the traffic sign recognition apparatus). The processor 40 executes various functional applications of the device and data processing by executing software programs, instructions and modules stored in the memory 41, that is, implements the above-described traffic sign recognition method. The method comprises the following steps:
acquiring pictures in an identification picture group to identify large classes of traffic signs, and acquiring a target picture of which an identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the large classes of the traffic signs identified in the identification region and an identification confidence coefficient;
screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture;
and performing traffic sign fine classification recognition in the at least one target judgment area, and acquiring traffic sign fine classification which meets a second confidence degree condition as a traffic sign recognition result.
The memory 41 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 41 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 41 may further include memory located remotely from processor 40, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 42 is operable to receive input numeric or character information and to generate key signal inputs relating to user settings and function controls of the apparatus. The output device 43 may include a display device such as a display screen.
EXAMPLE five
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, perform a method for identifying a traffic sign, the method including:
acquiring pictures in an identification picture group to identify large classes of traffic signs, and acquiring a target picture of which an identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the large classes of the traffic signs identified in the identification region and an identification confidence coefficient;
screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture;
and performing traffic sign fine classification recognition in the at least one target judgment area, and acquiring traffic sign fine classification which meets a second confidence degree condition as a traffic sign recognition result.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the traffic sign identification method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the traffic sign identifier, the included units and modules are merely divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A traffic sign recognition method, comprising:
acquiring pictures in an identification picture group to identify large classes of traffic signs, and acquiring a target picture of which an identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the large classes of the traffic signs identified in the identification region and an identification confidence coefficient;
screening out at least one target judgment area meeting the traffic sign identification condition in each identification area of the target picture;
and performing traffic sign fine classification recognition in the at least one target judgment area, and acquiring traffic sign fine classification which meets a second confidence degree condition as a traffic sign recognition result.
2. The method of claim 1, wherein obtaining pictures in the recognition picture group for traffic sign large category recognition, and obtaining the target picture whose recognition result meets the first confidence condition comprises:
acquiring a picture in the middle position from the identification picture group as a basic picture;
carrying out traffic sign large category identification on the basic picture, and acquiring the traffic sign large category and the identification confidence coefficient identified in each identification area;
and if the recognition confidence of the basic picture exceeds a preset first confidence threshold, respectively performing traffic sign large category recognition on the rest pictures in the recognition picture group, and acquiring the picture with the highest recognition confidence as the target picture.
3. The method according to claim 1, wherein screening out at least one target determination region satisfying a traffic sign recognition condition in each of the recognition regions of the target picture comprises at least one of:
if the length-width ratio of the traffic sign contained in the target identification area is determined to be abnormal, deleting the target identification area;
if the sum of the lengths and the widths of the traffic signs contained in the target identification area is smaller than the sum of the preset lengths and widths, deleting the target identification area; and
and if the traffic sign in the target identification area in which the forbidden traffic sign is identified is determined not to meet the forbidden traffic sign condition, deleting the target identification area.
4. The method of claim 3, wherein determining that the traffic sign within the target identification area in which the no-entry traffic sign is identified does not satisfy a no-entry traffic sign condition comprises:
converting the target identification area into a hexagonal pyramid model format;
counting the number of red pixel points in the target identification area according to the hue parameters of the pixel points;
and if the number of the red pixel points is less than the preset number of pixels, determining that the traffic sign contained in the target identification area does not meet the forbidden traffic sign condition.
5. The method of claim 1, wherein performing fine traffic sign category identification in the at least one target judgment area, and obtaining a fine traffic sign category satisfying a second confidence condition as a traffic sign identification result, comprises:
performing traffic sign fine classification identification on the at least one target judgment area to obtain an identification result, wherein the identification result comprises the traffic sign fine classification identified in the target judgment area and an identification confidence coefficient;
and if the recognition confidence of the target judgment area exceeds a preset second confidence threshold, the recognized traffic sign fine classification meets a second confidence condition, and the recognized traffic sign fine classification is used as a traffic sign recognition result.
6. The method according to any one of claims 1-5, wherein before obtaining the pictures in the group of identification pictures for traffic sign large category identification, the method further comprises:
cutting a road picture shot in real time, and storing the cut road picture into a picture queue;
acquiring a set number of road pictures in the picture queue as the identification picture group;
after acquiring the recognition result meeting the second confidence condition as the traffic sign recognition result, the method further comprises the following steps:
and returning to execute the operation of acquiring the road pictures with the set number in the picture queue as the identification picture group until the identification finishing condition is met.
7. The method of claim 6, wherein cropping the road picture taken in real-time comprises:
and cutting the shot road picture into a road picture with a preset size, wherein the cut road picture is the upper right part of the shot road picture.
8. A traffic sign recognition apparatus, comprising:
the traffic sign large category identification module is used for acquiring pictures in the identification picture group to identify the traffic sign large category and acquiring a target picture of which the identification result meets a first confidence coefficient condition, wherein the identification result comprises at least one identification region, the traffic sign large category identified in the identification region and an identification confidence coefficient;
the identification condition screening module is used for screening out at least one target judgment area meeting the identification condition of the traffic sign in each identification area of the target picture;
and the traffic sign fine classification identification module is used for carrying out traffic sign fine classification identification in the at least one target judgment area and acquiring the traffic sign fine classification meeting a second confidence coefficient condition as a traffic sign identification result.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the traffic sign recognition method according to any one of claims 1-7 when executing the program.
10. A storage medium containing computer-executable instructions for performing the traffic sign recognition method of any one of claims 1-7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911149014.9A CN110909674B (en) | 2019-11-21 | 2019-11-21 | Traffic sign recognition method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911149014.9A CN110909674B (en) | 2019-11-21 | 2019-11-21 | Traffic sign recognition method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110909674A true CN110909674A (en) | 2020-03-24 |
| CN110909674B CN110909674B (en) | 2024-01-05 |
Family
ID=69818301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911149014.9A Active CN110909674B (en) | 2019-11-21 | 2019-11-21 | Traffic sign recognition method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110909674B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111582029A (en) * | 2020-04-02 | 2020-08-25 | 天津大学 | Traffic sign identification method based on dense connection and attention mechanism |
| CN112699841A (en) * | 2021-01-13 | 2021-04-23 | 华南理工大学 | Traffic sign detection and identification method based on driving video |
| CN113963329A (en) * | 2021-10-11 | 2022-01-21 | 浙江大学 | Digital traffic sign detection and identification method based on double-stage convolutional neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809121A (en) * | 2016-03-03 | 2016-07-27 | 电子科技大学 | Multi-characteristic synergic traffic sign detection and identification method |
| CN107220646A (en) * | 2017-05-25 | 2017-09-29 | 杭州健培科技有限公司 | A kind of medical image Text region Enhancement Method for going ambient interferences |
| CN109255279A (en) * | 2017-07-13 | 2019-01-22 | 深圳市凯立德科技股份有限公司 | A kind of method and system of road traffic sign detection identification |
-
2019
- 2019-11-21 CN CN201911149014.9A patent/CN110909674B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809121A (en) * | 2016-03-03 | 2016-07-27 | 电子科技大学 | Multi-characteristic synergic traffic sign detection and identification method |
| CN107220646A (en) * | 2017-05-25 | 2017-09-29 | 杭州健培科技有限公司 | A kind of medical image Text region Enhancement Method for going ambient interferences |
| CN109255279A (en) * | 2017-07-13 | 2019-01-22 | 深圳市凯立德科技股份有限公司 | A kind of method and system of road traffic sign detection identification |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111582029A (en) * | 2020-04-02 | 2020-08-25 | 天津大学 | Traffic sign identification method based on dense connection and attention mechanism |
| CN111582029B (en) * | 2020-04-02 | 2022-08-12 | 天津大学 | Traffic sign identification method based on dense connection and attention mechanism |
| CN112699841A (en) * | 2021-01-13 | 2021-04-23 | 华南理工大学 | Traffic sign detection and identification method based on driving video |
| CN113963329A (en) * | 2021-10-11 | 2022-01-21 | 浙江大学 | Digital traffic sign detection and identification method based on double-stage convolutional neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110909674B (en) | 2024-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11417118B2 (en) | Lane line data processing method and apparatus, computer device, and storage medium | |
| CN109087510B (en) | Traffic monitoring method and device | |
| CN110059694B (en) | Intelligent identification method for character data in complex scene of power industry | |
| US11455805B2 (en) | Method and apparatus for detecting parking space usage condition, electronic device, and storage medium | |
| US11176418B2 (en) | Model test methods and apparatuses | |
| CN107944450B (en) | License plate recognition method and device | |
| CN111784685A (en) | Power transmission line defect image identification method based on cloud edge cooperative detection | |
| KR20210078530A (en) | Lane property detection method, device, electronic device and readable storage medium | |
| CN110909674A (en) | Traffic sign identification method, device, equipment and storage medium | |
| CN112329881B (en) | License plate recognition model training method, license plate recognition method and device | |
| CN112819068B (en) | Ship operation violation behavior real-time detection method based on deep learning | |
| CN110599453A (en) | Panel defect detection method and device based on image fusion and equipment terminal | |
| WO2021185121A1 (en) | Model generation method and apparatus, object detection method and apparatus, device, and storage medium | |
| CN112949578B (en) | Vehicle lamp state identification method, device, equipment and storage medium | |
| CN108229473A (en) | Vehicle annual inspection label detection method and device | |
| CN112906471A (en) | Traffic signal lamp identification method and device | |
| CN110705338A (en) | Vehicle detection method and device and monitoring equipment | |
| CN116052090A (en) | Image quality evaluation method, model training method, device, equipment and medium | |
| CN111046747B (en) | Crowd counting model training method, crowd counting method, device and server | |
| CN112001453A (en) | Method and device for calculating accuracy of video event detection algorithm | |
| CN114429640A (en) | Drawing segmentation method and device and electronic equipment | |
| CN112241659A (en) | Detection device and detection method for illegal building and terminal equipment | |
| CN113435350A (en) | Traffic marking detection method, device, equipment and medium | |
| CN116994721B (en) | Quick processing system of digital pathological section graph | |
| CN111985439B (en) | Face detection method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |