CN112016503A - Sidewalk detection method and device, computer equipment and storage medium - Google Patents

Sidewalk detection method and device, computer equipment and storage medium Download PDF

Info

Publication number
CN112016503A
CN112016503A CN202010923999.2A CN202010923999A CN112016503A CN 112016503 A CN112016503 A CN 112016503A CN 202010923999 A CN202010923999 A CN 202010923999A CN 112016503 A CN112016503 A CN 112016503A
Authority
CN
China
Prior art keywords
sidewalk
image
detection model
loss
processed
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
Application number
CN202010923999.2A
Other languages
Chinese (zh)
Other versions
CN112016503B (en
Inventor
吴晓东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An International Smart City Technology Co Ltd
Original Assignee
Ping An International Smart City Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An International Smart City Technology Co Ltd filed Critical Ping An International Smart City Technology Co Ltd
Priority to CN202010923999.2A priority Critical patent/CN112016503B/en
Publication of CN112016503A publication Critical patent/CN112016503A/en
Application granted granted Critical
Publication of CN112016503B publication Critical patent/CN112016503B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • 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
    • G06V10/32—Normalisation of the pattern dimensions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Probability & Statistics with Applications (AREA)
  • Image Analysis (AREA)

Abstract

The embodiment of the application belongs to the technical field of artificial intelligence, is applied to the field of intelligent traffic, and relates to a sidewalk detection method, which comprises the steps of receiving and adjusting a training image to be a preset size as an image to be processed; detecting an image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, wherein the sidewalk feature extraction network comprises convolution, batch normalization and activation; calculating a loss function of the detection model according to the first sidewalk coordinate, and updating and iterating the detection model until the detection model converges to obtain a trained detection model; and receiving an original image, inputting the original image into the trained detection model to obtain a second sidewalk coordinate, and mapping the second sidewalk coordinate to be a coordinate on the original image. The application also provides a sidewalk detection device, computer equipment and a storage medium. The trained detection model may be stored in a blockchain. The method and the device effectively improve the accuracy of the computer for detecting the sidewalk.

Description

Sidewalk detection method and device, computer equipment and storage medium
Technical Field
The present application relates to the field of artificial intelligence technologies, and in particular, to a sidewalk detection method and apparatus, a computer device, and a storage medium.
Background
The sidewalk is an important road sign for guaranteeing the personal safety of pedestrians, and the automatic detection and identification of the sidewalk are important technical links in an intelligent traffic system. The deep learning method based on YOLOv3 becomes one of popular sidewalk detection algorithms in the industry at present due to the high detection speed.
At present, a sidewalk detection algorithm based on YOLOv3 can achieve the effect of real-time detection and high accuracy rate in simple scenes such as sunny days, daytime and clear sidewalk zebra stripes, but has low accuracy rate and recall rate in difficult scenes such as haze, rainy days, nighttime and fuzzy sidewalk zebra stripes, and is difficult to effectively detect and identify the sidewalk.
Disclosure of Invention
The embodiment of the application aims to provide a sidewalk detection method, a sidewalk detection device, a computer device and a storage medium, and the accuracy of the sidewalk detection by a computer is obviously improved.
In order to solve the above technical problem, an embodiment of the present application provides a sidewalk detection method, which adopts the following technical solutions:
a method of walkway inspection comprising the steps of:
receiving a training image, and adjusting the size of the training image to be a preset size as an image to be processed;
detecting the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, wherein the sidewalk feature extraction network comprises convolution, batch normalization and activation;
calculating a loss function of the detection model according to the first sidewalk coordinate, and performing updating iteration on the detection model until the detection model is converged to obtain a trained detection model; and
and receiving an original image, inputting the original image into the trained detection model to obtain a second sidewalk coordinate, and mapping the second sidewalk coordinate to be a coordinate on the original image.
Further, the step of detecting the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate includes:
performing feature extraction on the image to be processed through a preset sidewalk feature extraction network to obtain sidewalk features;
performing CBL, convolution and up-sampling operations in the detection model on the sidewalk feature to obtain a first feature map, a second feature map and a third feature map;
and predicting the sidewalk for the first feature map, the second feature map and the third feature map based on a k-means algorithm to obtain a first sidewalk coordinate.
Further, the step of extracting the features of the image to be processed through a preset sidewalk feature extraction network to obtain the sidewalk features includes:
and sequentially executing CBL and CResX in a pedestrian path feature extraction network on the image to be processed to obtain the pedestrian path features in the image to be processed.
Further, the CResX includes ResX, a first CBL, and a second CBL, and the step of sequentially performing a pedestrian path feature extraction network on the to-be-processed image to extract the CBL and the CResX in the network to obtain the pedestrian path feature in the to-be-processed image includes:
executing CBL in a pedestrian path feature extraction network on the image to be processed to obtain a first intermediate image;
sequentially passing the first intermediate image through ResX and first CBL of CResX to obtain a second intermediate image;
the first intermediate image passes through a second CBL of CResX to obtain a third intermediate image;
and splicing the second intermediate image and the third intermediate image to obtain the sidewalk characteristics in the image to be processed.
Further, the step of sequentially performing CBL and CResX in the sidewalk feature extraction network on the image to be processed to obtain the sidewalk feature in the image to be processed includes:
and sequentially executing CBL, CRes1, CRes2, CRes8, CRes8 and CRes4 in a sidewalk feature extraction network on the image to be processed to obtain the sidewalk feature in the image to be processed.
Further, the step of detecting the image to be processed by using a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate includes:
detecting the image to be processed by using a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate and a first sidewalk category; and is
The step of calculating a loss function of the detection model from the first sidewalk coordinates comprises:
taking the first sidewalk coordinate as a coordinate of a sidewalk prediction frame, obtaining category loss based on the category, respectively obtaining center point coordinate loss, width and height loss and DIOU loss of the sidewalk prediction frame based on the coordinate of the sidewalk prediction frame, and calculating a loss function of the detection model according to the following formula:
loss＝∑(lossxy+losswh+lossclass+lossdiou)
therein, lossxyLoss of coordinates of center point of prediction frame for pedestrian pathwhLoss of width for pedestrian path prediction boxclassLoss of classdiouIs a DIOU loss.
Further, the calculation formula of the DIOU loss is as follows:
Figure BDA0002667699320000031
wherein d (box _ predict, box _ true) represents the distance between the center points of the sidewalk prediction frame and the sidewalk real labeling frame, c represents the diagonal length of the minimum circumscribed matrix surrounding the sidewalk prediction frame and the sidewalk real labeling frame at the same time, and IOU represents the intersection and parallel ratio between the areas of the sidewalk prediction frame and the sidewalk real labeling frame.
In order to solve the above technical problem, an embodiment of the present application further provides a sidewalk detection device, which adopts the following technical scheme:
a pavement detection apparatus comprising:
the adjusting module is used for receiving a training image, adjusting the size of the training image to be a preset size and taking the training image as an image to be processed;
the detection module is used for detecting the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, wherein the sidewalk feature extraction network comprises convolution, batch normalization and activation;
the iteration module is used for calculating a loss function of the detection model according to the first sidewalk coordinate, updating and iterating the detection model until the detection model is converged, and obtaining a trained detection model; and
and the obtaining module is used for receiving the original image, inputting the original image into the trained detection model, obtaining a second sidewalk coordinate, and mapping the second sidewalk coordinate to a coordinate on the original image.
In order to solve the above technical problem, an embodiment of the present application further provides a computer device, which adopts the following technical solutions:
a computer device comprising a memory having computer readable instructions stored therein and a processor that when executed performs the steps of the method for detecting a sidewalk described above.
In order to solve the above technical problem, an embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solutions:
a computer readable storage medium having computer readable instructions stored thereon which, when executed by a processor, implement the steps of the sidewalk detection method described above.
Compared with the prior art, the embodiment of the application mainly has the following beneficial effects:
according to the method and the device, the size of the training image is adjusted, so that the obtained to-be-processed image meets the input requirement of the detection model, the accuracy of the detection model identification is improved, and the situation that the accuracy of the detection model identification is reduced due to the fact that the training image is directly input into the model is avoided. The feature expression capability of the sidewalk feature extraction network is improved through convolution, batch normalization and activation in the sidewalk feature extraction network, and therefore the accuracy of the detection model for identifying the sidewalks is improved. And through calculation of the loss function and iteration of the detection model, convergence of the detection model is realized, and the trained detection model is obtained.
Drawings
In order to more clearly illustrate the solution of the present application, the drawings needed for describing the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and that other drawings can be obtained by those skilled in the art without inventive effort.
FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;
FIG. 2 is a flow diagram of one embodiment of a sidewalk detection method according to the present application;
FIG. 3 is a schematic diagram of the structure of one embodiment of a pavement detection apparatus according to the present application;
FIG. 4 is a schematic block diagram of one embodiment of a computer device according to the present application.
Reference numerals: 200. a computer device; 201. a memory; 202. a processor; 203. a network interface; 300. a sidewalk detection device; 301. an adjustment module; 302. a detection module; 303. an iteration module; 304. a module is obtained.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "including" and "having," and any variations thereof, in the description and claims of this application and the description of the above figures are intended to cover non-exclusive inclusions. The terms "first," "second," and the like in the description and claims of this application or in the above-described drawings are used for distinguishing between different objects and not for describing a particular order.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, mpeg compression standard Audio Layer 3), MP4 players (Moving Picture Experts Group Audio Layer IV, mpeg compression standard Audio Layer 4), laptop portable computers, desktop computers, and the like.
The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.
It should be noted that the sidewalk detection method provided in the embodiments of the present application is generally executed by a server/terminal device, and accordingly, the sidewalk detection apparatus is generally disposed in the server/terminal device.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to fig. 2, a flow diagram of one embodiment of a pavement detection method according to the present application is shown. The sidewalk detection method comprises the following steps:
s1: receiving a training image, and adjusting the size of the training image to be a preset size as an image to be processed.
In this embodiment, the input training image is resized to a fixed size of 512 × 512 pixels to meet the requirements of the sidewalk feature extraction network.
In this embodiment, the electronic device (e.g., the server/terminal device shown in fig. 1) on which the sidewalk detection method operates may receive the training image or the original image through a wired connection or a wireless connection. It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra wideband) connection, and other wireless connection means now known or developed in the future.
S2: and detecting the image to be processed by a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, wherein the sidewalk feature extraction network comprises convolution, batch normalization and activation.
In the present embodiment, the yollov 3 algorithm includes a DarkNet53 network, and the present application is directed to a DarkNet53 network in yollov 3. It is improved into the sidewalk characteristic extraction network of the application, namely a DarkNet63 network. The DarkNet63 network obviously improves the feature expression capability of the network, and effectively improves the feature extraction quality of sidewalks in difficult scenes such as haze, rainy days, nights, pedestrian crossing and the like, thereby improving the overall accuracy and recall rate of pedestrian detection.
Specifically, in step S2, the step of obtaining the first sidewalk coordinate by detecting the image to be processed through the detection model based on the sidewalk feature extraction network includes:
performing feature extraction on the image to be processed through a preset sidewalk feature extraction network to obtain sidewalk features;
performing operations of CBL, convolution and up-sampling preset in the detection model on the sidewalk feature to obtain a first feature map, a second feature map and a third feature map;
and predicting the sidewalk for the first feature map, the second feature map and the third feature map based on a k-means algorithm to obtain a first sidewalk coordinate.
In this embodiment, the CBL (i.e., convolution Conv + batch normalized BN + activation function leak relu), Conv (convolution), and Upsample (up-sampling) are performed multiple times to obtain three feature maps of different scales, i.e., a first feature map y1, a second feature map y2, and a third feature map y 3; the operation of the feature maps y2 and y3 is mainly one residual connecting operation more than that of y1, namely, the feature map obtained at the previous layer is subjected to up-sampling and then spliced with the corresponding block in the DarkNet63 network. The implementation process of performing CBL, convolution and upsampling operations in the detection model on the sidewalk feature in the present application is consistent with the corresponding implementation process in YOLOv3 algorithm, and will not be described herein too much. And calling 9 anchor boxes (rectangular boxes) obtained by clustering through a k-means algorithm in advance to detect and identify the pedestrian lanes on the three feature graphs with different scales, namely a first feature graph, a second feature graph and a third feature graph (y1, y2 and y3), and predicting the coordinates and the categories (namely two categories, namely whether the pedestrian lanes are) of the 3 different anchor boxes (rectangular boxes) on each feature graph respectively. And processing the prediction result of the rectangular frame based on a non-maximum suppression (NMS) algorithm to obtain a first sidewalk coordinate. And more accurate sidewalk coordinates are obtained, so that the accuracy of the detection model is improved.
The method comprises the following steps of carrying out feature extraction on the image to be processed through a preset sidewalk feature extraction network, wherein the step of obtaining the sidewalk features comprises the following steps:
and sequentially executing CBL and CResX in a pedestrian path feature extraction network on the image to be processed to obtain the pedestrian path features in the image to be processed.
In the present embodiment, the CBL includes Conv (convolution), BN (batch normalization), leak relu (activation function). And obtaining a deeper feature map, namely sidewalk features, through CBL and CResX, wherein the feature expression capability of the sidewalk feature extraction network is improved by adding an activation function.
Further, the CResX includes ResX, a first CBL, and a second CBL, the step of sequentially performing a pedestrian path feature extraction network of CBL and CResX on the image to be processed to obtain the pedestrian path feature in the image to be processed includes:
executing CBL in a pedestrian path feature extraction network on the image to be processed to obtain a first intermediate image;
sequentially passing the first intermediate image through ResX and first CBL of CResX to obtain a second intermediate image;
the first intermediate image passes through a second CBL of CResX to obtain a third intermediate image;
and splicing the second intermediate image and the third intermediate image to obtain the sidewalk characteristics in the image to be processed.
In this embodiment, the second intermediate image and the third intermediate image are stitched to represent two matrix stitching, that is, the matrix size is not changed, and the number of channels is changed. For example: the input matrix size of the CResX is 60 × 128, that is, the matrix size obtained after the ResX + first CBL operation is 60 × 200, and the matrix size obtained after the second CBL operation is 60 × 300, and then the matrix size obtained after the two matrices are subjected to the concatenation (Concat) operation is 60 × 60 (200+300) × 60 × 500. Where Resx includes CBL and X Res units. X represents a number, Res1, Res2, …, Res8, etc., indicating how many Res _ units are contained in this Res. And inputting the image characteristics into ResX, and outputting after CBL and X Res units. Where Res unit is to perform CBL twice on the image features in Res unit, and Add the image features obtained after performing CBL twice to the image features in Res unit originally input (i.e., Add, matrix addition). The accuracy of image processing is refined, images obtained through different processing modes are spliced to serve as sidewalk features, and the feature expression capability of the detection model is enhanced.
In addition, the step of sequentially executing CBL and CResX in the sidewalk feature extraction network on the image to be processed to obtain the sidewalk feature in the image to be processed further includes:
and sequentially executing CBL, CRes1, CRes2, CRes8, CRes8 and CRes4 in a sidewalk feature extraction network on the image to be processed to obtain the sidewalk feature in the image to be processed.
In this embodiment, the DarkNet63 network of the present application includes CBL, crs 1, crs 2, crs 8, crs 8, and crs 4, where crs 1 means that there are 1 Res _ unit, crs 2 means that there are 2 Res _ units, crs 8 means that there are 8 Res _ units, and crs 4 means that there are 4 Res _ units. And the image to be processed is sequentially processed by the different CResX, so that the obtained sidewalk characteristics are more accurate.
S3: calculating a loss function of the detection model according to the first sidewalk coordinate, and performing updating iteration on the detection model until the detection model is converged to obtain a trained detection model;
in this embodiment, a loss between a prediction result (sidewalk prediction frame coordinates and categories) and a labeling result (sidewalk real labeling frame coordinates and categories) is calculated by using a loss function, and finally, updating iteration of model parameters is performed by using a random gradient descent algorithm with the minimized loss as an optimization target, so that model convergence is finally achieved, that is, model training is completed.
Specifically, the step of detecting the image to be processed by using a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate includes:
detecting the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate and a first sidewalk category;
the step of calculating a loss function of the detection model from the first sidewalk coordinates comprises:
taking the first sidewalk coordinate as a coordinate of a sidewalk prediction frame, obtaining category loss based on the category, respectively obtaining center point coordinate loss, width and height loss and DIOU loss of the sidewalk prediction frame based on the coordinate of the sidewalk prediction frame, and calculating a loss function of the detection model according to the following formula:
loss＝∑(lossxy+losswh+lossclass+lossdiou)
therein, lossxyLoss of coordinates of center point of prediction frame for pedestrian pathwhLoss of width for pedestrian path prediction boxclassLoss of classdiouIs a DIOU loss.
In this embodiment, the loss function of the model consists of three parts, namely, coordinate loss, category loss and DIOU loss, wherein the coordinate loss includes two parts, namely, coordinate loss of the center point of the sidewalk prediction frame and width and height loss of the sidewalk prediction frame. According to the method and the device, the original loss function is improved, the calculation of the existing loss function only comprises coordinate loss and category loss, the DIOU loss is increased on the basis, the position and the size of the sidewalk detection frame can be fitted more accurately, and therefore the overall accuracy and the recall rate of sidewalk detection are further improved.
The calculation formula of the coordinate loss of the center point of the sidewalk prediction frame is as follows:
Figure BDA0002667699320000102
wherein, ytrue_xValue y of the abscissa representing the center point of the real marking frame of the lanepredjct_xValue of the abscissa, y, which is the center point of the pedestrian path prediction boxtrue_yValue y of the ordinate of the center point of the real marking frame of the person pathpredict_yA value of a vertical coordinate of a central point of the pedestrian path prediction box;
the calculation formula of the width and height loss of the sidewalk prediction frame is as follows:
Figure BDA0002667699320000101
wherein, ytrue_wValue y of width of real mark box for pedestrian pathpredcit_wPredicting the value of the width of the box, y, for the sidewalktrue_hHigh value, y, for the real mark box of the lanepredict_hPredict a high value of the box for the sidewalk;
the category loss is calculated as follows:
lossclass＝-(ytrue_classlogypredict_class+(1-ytrue_class)log(1-ypredict_class))
wherein, ytrue_classClass of real mark box for pedestrian path, ypredict_classPredicting a category of a box for the sidewalk;
the calculation formula of the DIOU loss is as follows:
Figure BDA0002667699320000111
d (box _ predict, box _ true) represents the distance between the central points of the sidewalk prediction frame and a sidewalk real labeling frame labeled on the training image in advance, and c represents the diagonal length of a minimum circumscribed matrix surrounding the sidewalk prediction frame and the sidewalk real labeling frame at the same time; the IOU is the intersection ratio between the areas of the pedestrian path prediction frame and the real sidewalk marking frame.
In this embodiment, the calculation formula of the IOU of the present application is:
Figure BDA0002667699320000112
wherein, I represents the intersection area of the sidewalk prediction frame and the sidewalk real marking frame, and U represents the sidewalk prediction frame and the sidewalk real marking frameNote the union area of the boxes. The iou (intersection over union) is the result of dividing the overlapping portion of two regions by the aggregation portion of the two regions. Loss of center point coordinates (loss) of sidewalk prediction framexy) Is the sum of the cross entropy losses for coordinate x and coordinate y. Loss of width of sidewalk prediction framewh) Is the sum of the mean square losses of width w and height h. Loss of class (loss)class) Is the cross entropy loss of the class. DIOU loss (loss)diou) And predicting the loss between the boxes and the marking boxes for the sidewalks. Through DIOU (Distance-IoU), the better regression prediction can be carried out on the anchors box with different proportions, areas and directions.
It should be noted that: the sidewalk real labeling frame in the application is a labeling frame labeled on a training image in advance. The content in its box is the sidewalk image in the training image, which represents the true position of the sidewalk image. The method and the device have the advantages that the coordinates, the coordinates of the central point, the width, the height, the length, the category and the like of each vertex in the real sidewalk labeling frame are labeled in advance.
S4: and receiving an original image, inputting the original image into the trained detection model to obtain a second sidewalk coordinate, and mapping the second sidewalk coordinate to be a coordinate on the original image.
In this embodiment, the predicted sidewalk coordinates are mapped to coordinates on an original image, so as to implement sidewalk detection of the image, wherein after the original image is input into a detection model, the first feature map, the second feature map and the third feature map output by the detection model are obtained, and after the three feature maps are predicted based on a k-means algorithm, the second sidewalk coordinates are obtained.
According to the method and the device, the size of the training image is adjusted, so that the obtained to-be-processed image meets the input requirement of the detection model, the accuracy of the detection model identification is improved, and the situation that the accuracy of the detection model identification is reduced due to the fact that the training image is directly input into the model is avoided. The feature expression capability of the sidewalk feature extraction network is improved through convolution, batch normalization and activation in the sidewalk feature extraction network, and therefore the accuracy of the detection model for identifying the sidewalks is improved. And through calculation of the loss function and iteration of the detection model, convergence of the detection model is realized, and the trained detection model is obtained.
It is emphasized that, in order to further ensure the privacy and security of the trained detection model, the trained detection model may also be stored in a node of a blockchain.
The block chain referred by the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The application can be applied to the field of intelligent traffic, and therefore the construction of an intelligent city is promoted.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware associated with computer readable instructions, which can be stored in a computer readable storage medium, and when executed, can include processes of the embodiments of the methods described above. The storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless explicitly stated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
With further reference to fig. 3, as an implementation of the method shown in fig. 2, the present application provides an embodiment of a sidewalk detection apparatus, which corresponds to the embodiment of the method shown in fig. 2, and which can be applied to various electronic devices.
As shown in fig. 3, the sidewalk detection apparatus 300 according to the present embodiment includes: an adjustment module 301, a detection module 302, an iteration module 303, and an obtaining module 304. Wherein: the adjusting module 301 is configured to receive a training image, adjust the size of the training image to a preset size, and use the size as an image to be processed; the detection module 302 is configured to detect the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, where the sidewalk feature extraction network includes convolution, batch normalization, and activation; the iteration module 303 is configured to calculate a loss function of the detection model according to the first sidewalk coordinate, and perform update iteration on the detection model until the detection model converges to obtain a trained detection model; the obtaining module 304 is configured to receive an original image, input the original image into the trained detection model, obtain a second sidewalk coordinate, and map the second sidewalk coordinate to a coordinate on the original image.
In the embodiment, the size of the training image is adjusted, so that the obtained image to be processed meets the input requirement of the detection model, the accuracy of the detection model identification is improved, and the situation that the accuracy of the detection model identification is reduced due to the fact that the training image is directly input into the model is avoided. The feature expression capability of the sidewalk feature extraction network is improved through convolution, batch normalization and activation in the sidewalk feature extraction network, and therefore the accuracy of the detection model for identifying the sidewalks is improved. And through calculation of the loss function and iteration of the detection model, convergence of the detection model is realized, and the trained detection model is obtained.
The detection module 302 includes an extraction sub-module, an operation sub-module, and a prediction sub-module. The extraction submodule is used for extracting the features of the image to be processed through a preset sidewalk feature extraction network to obtain sidewalk features; the operation submodule is used for executing CBL, convolution and up-sampling operations preset in the detection model on the sidewalk characteristics to obtain a first characteristic diagram, a second characteristic diagram and a third characteristic diagram, and the prediction submodule is used for predicting the sidewalk on the basis of a k-means algorithm on the first characteristic diagram, the second characteristic diagram and the third characteristic diagram to obtain a first sidewalk coordinate.
In some optional implementations of this embodiment, the extracting sub-module is further configured to: and sequentially executing CBL and CResX in a pedestrian path feature extraction network on the image to be processed to obtain the pedestrian path features in the image to be processed.
The CResX comprises a ResX, a first CBL and a second CBL, and the extraction submodule comprises a first execution unit, a second execution unit, a third execution unit and a splicing unit. The first execution unit is used for executing CBL in the pedestrian path feature extraction network on the image to be processed to obtain a first intermediate image; the second execution unit is used for enabling the first intermediate image to sequentially pass through ResX and first CBL of CResX to obtain a second intermediate image; the third execution unit is used for enabling the first intermediate image to pass through a second CBL of CResX to obtain a third intermediate image; and the splicing unit is used for splicing the second intermediate image and the third intermediate image to obtain the sidewalk characteristics in the image to be processed.
In some optional implementations of this embodiment, the extracting sub-module is further configured to: and sequentially executing CBL, CRes1, CRes2, CRes8, CRes8 and CRes4 in a sidewalk feature extraction network on the image to be processed to obtain the sidewalk feature in the image to be processed.
In some optional implementation manners of this embodiment, the detection module is further configured to detect the image to be processed through a detection model based on a sidewalk feature extraction network, so as to obtain a first sidewalk coordinate and a first category; the iteration module is further configured to use the first sidewalk coordinate as a coordinate of a sidewalk prediction frame, obtain a category loss based on the category, respectively obtain a center point coordinate loss of the sidewalk prediction frame, a width and height loss of the sidewalk prediction frame, and a DIOU loss based on the coordinate of the sidewalk prediction frame, and calculate a loss function of the detection model according to the following formula:
loss＝∑(lossxy+losswh+lossclass+lossdiou)
therein, lossxyLoss of coordinates of center point of prediction frame for pedestrian pathwhLoss of width for pedestrian path prediction boxclassLoss of classdiouIs a DIOU loss.
The iteration module includes a center point sub-module, a width-height sub-module, a category sub-module, and a loss sub-module. The central point submodule is used for calculating the coordinate loss of the central point of the sidewalk prediction frame, and the formula is as follows:
Figure BDA0002667699320000141
wherein, ytrue_xValue y of the abscissa representing the center point of the real marking frame of the lanepredict_xValue of the abscissa, y, which is the center point of the pedestrian path prediction boxtrue_yValue y of the ordinate of the center point of the real marking frame of the person pathpredict_yIs the value of the ordinate of the center point of the pedestrian path prediction box.
The width and height sub-module is used for calculating the width and height loss of the sidewalk prediction frame, and the formula is as follows:
Figure BDA0002667699320000151
wherein, ytrue_wValue y of width of real mark box for pedestrian pathpredcit_wPredicting the value of the width of the box, y, for the sidewalktrue_hHigh value, y, for the real mark box of the lanepredict_hA high value of the box is predicted for the sidewalk.
The category submodule is used for calculating category loss, and the formula is as follows:
lossclass＝-(ytrue_classlogypredict_class+(1-ytrue_class)log(1-ypredict_class))
wherein, ytrue_classClass of real mark box for pedestrian path, ypredict_classThe category of the box is predicted for the sidewalk.
The loss submodule is used for calculating the DIOU loss, and the formula is as follows:
Figure BDA0002667699320000152
wherein d (box _ predict, box _ true) represents the distance between the central points of the sidewalk prediction frame and the sidewalk real labeling frame, and c represents the diagonal length of the minimum circumscribed matrix surrounding the sidewalk prediction frame and the sidewalk real labeling frame at the same time; the IOU is the intersection ratio between the areas of the pedestrian path prediction frame and the real sidewalk marking frame.
According to the method and the device, the size of the training image is adjusted, so that the obtained to-be-processed image meets the input requirement of the detection model, the accuracy of the detection model identification is improved, and the situation that the accuracy of the detection model identification is reduced due to the fact that the training image is directly input into the model is avoided. The feature expression capability of the sidewalk feature extraction network is improved through convolution, batch normalization and activation in the sidewalk feature extraction network, and therefore the accuracy of the detection model for identifying the sidewalks is improved. And through calculation of the loss function and iteration of the detection model, convergence of the detection model is realized, and the trained detection model is obtained.
In order to solve the technical problem, an embodiment of the present application further provides a computer device. Referring to fig. 4, fig. 4 is a block diagram of a basic structure of a computer device according to the present embodiment.
The computer device 200 comprises a memory 201, a processor 202, a network interface 203 communicatively connected to each other via a system bus. It is noted that only computer device 200 having components 201 and 203 is shown, but it is understood that not all of the illustrated components are required and that more or fewer components may alternatively be implemented. As will be understood by those skilled in the art, the computer device is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.
The computer device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The computer equipment can carry out man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch panel or voice control equipment and the like.
The memory 201 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the storage 201 may be an internal storage unit of the computer device 200, such as a hard disk or a memory of the computer device 200. In other embodiments, the memory 201 may also be an external storage device of the computer device 200, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like, provided on the computer device 200. Of course, the memory 201 may also include both internal and external storage devices of the computer device 200. In this embodiment, the memory 201 is generally used for storing an operating system installed in the computer device 200 and various types of application software, such as computer readable instructions of a sidewalk detection method. Further, the memory 201 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 202 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 202 is generally operative to control overall operation of the computer device 200. In this embodiment, the processor 202 is configured to execute computer readable instructions stored in the memory 201 or process data, such as computer readable instructions for executing the sidewalk detection method.
The network interface 203 may comprise a wireless network interface or a wired network interface, and the network interface 203 is generally used for establishing communication connection between the computer device 200 and other electronic devices.
In the embodiment, the novel sidewalk feature extraction network is adopted, so that the feature expression capability of the network is effectively enhanced, and the overall accuracy and recall rate of sidewalk detection are improved.
The present application further provides another embodiment, which is to provide a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of the sidewalk detection method as described above.
In the embodiment, the novel sidewalk feature extraction network is adopted, so that the feature expression capability of the network is effectively enhanced, and the overall accuracy and recall rate of sidewalk detection are improved.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.
It is to be understood that the above-described embodiments are merely illustrative of some, but not restrictive, of the broad invention, and that the appended drawings illustrate preferred embodiments of the invention and do not limit the scope of the invention. This application is capable of embodiments in many different forms and is provided for the purpose of enabling a thorough understanding of the disclosure of the application. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to one skilled in the art that the present application may be practiced without modification or with equivalents of some of the features described in the foregoing embodiments. All equivalent structures made by using the contents of the specification and the drawings of the present application are directly or indirectly applied to other related technical fields and are within the protection scope of the present application.

Claims (10)

1. A method of detecting a walkway, comprising the steps of:
receiving a training image, and adjusting the size of the training image to be a preset size as an image to be processed;
detecting the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, wherein the sidewalk feature extraction network comprises convolution, batch normalization and activation;
calculating a loss function of the detection model according to the first sidewalk coordinate, and performing updating iteration on the detection model until the detection model is converged to obtain a trained detection model; and
and receiving an original image, inputting the original image into the trained detection model to obtain a second sidewalk coordinate, and mapping the second sidewalk coordinate to be a coordinate on the original image.
2. The sidewalk detection method according to claim 1, wherein the step of obtaining the first sidewalk coordinates by detecting the image to be processed through a detection model based on a sidewalk feature extraction network comprises:
performing feature extraction on the image to be processed through a preset sidewalk feature extraction network to obtain sidewalk features;
performing operations of CBL, convolution and up-sampling preset in the detection model on the sidewalk feature to obtain a first feature map, a second feature map and a third feature map;
and predicting the sidewalk for the first feature map, the second feature map and the third feature map based on a k-means algorithm to obtain a first sidewalk coordinate.
3. The sidewalk detection method according to claim 2, wherein the step of performing feature extraction on the image to be processed through a preset sidewalk feature extraction network to obtain the sidewalk features comprises:
and sequentially executing CBL and CResX in the sidewalk feature extraction network on the image to be processed to obtain the sidewalk features in the image to be processed.
4. The sidewalk detection method according to claim 3, wherein the CResX comprises ResX, a first CBL and a second CBL, the CBL and the CResX in the sidewalk feature extraction network are sequentially performed on the image to be processed, and the step of obtaining the sidewalk features in the image to be processed comprises:
executing CBL in a pedestrian path feature extraction network on the image to be processed to obtain a first intermediate image;
sequentially passing the first intermediate image through ResX and first CBL of CResX to obtain a second intermediate image;
the first intermediate image passes through a second CBL of CResX to obtain a third intermediate image;
and splicing the second intermediate image and the third intermediate image to obtain the sidewalk characteristics in the image to be processed.
5. The sidewalk detection method according to claim 3, wherein the step of sequentially performing CBL and CResX in a sidewalk feature extraction network on the image to be processed to obtain the sidewalk features in the image to be processed further comprises:
and sequentially executing CBL, CRes1, CRes2, CRes8, CRes8 and CRes4 in a sidewalk feature extraction network on the image to be processed to obtain the sidewalk feature in the image to be processed.
6. The sidewalk detection method according to claim 1, wherein the step of detecting the image to be processed by using a detection model based on a sidewalk feature extraction network to obtain first sidewalk coordinates comprises:
detecting the image to be processed by using a detection model based on the sidewalk feature extraction network to obtain the coordinates and the category of the first sidewalk; and is
The step of calculating a loss function of the detection model from the first sidewalk coordinates comprises:
taking the first sidewalk coordinate as a coordinate of a sidewalk prediction frame, obtaining category loss based on the category, respectively obtaining center point coordinate loss, width and height loss and DIOU loss of the sidewalk prediction frame based on the coordinate of the sidewalk prediction frame, and calculating a loss function of the detection model according to the following formula:
loss＝∑(lossxy+losswh+lossclass+lossdiou)
therein, lossxyLoss of coordinates of center point of prediction frame for pedestrian pathwhLoss of width for pedestrian path prediction boxclassLoss of classdiouIs a DIOU loss.
7. The walkway detection method of claim 6, wherein the DIOU loss is calculated as follows:
Figure FDA0002667699310000021
wherein d (box _ predict, box _ true) represents a distance between a center point of the sidewalk prediction frame and a center point of a sidewalk real labeling frame labeled on a training image in advance, c represents a diagonal length of a minimum circumscribed matrix surrounding the sidewalk prediction frame and the sidewalk real labeling frame at the same time, and IOU represents an intersection-parallel ratio between areas of the sidewalk prediction frame and the sidewalk real labeling frame.
8. A pavement detection apparatus, comprising:
the adjusting module is used for receiving a training image, adjusting the size of the training image to be a preset size and taking the training image as an image to be processed;
the detection module is used for detecting the image to be processed through a detection model based on a sidewalk feature extraction network to obtain a first sidewalk coordinate, wherein the sidewalk feature extraction network comprises convolution, batch normalization and activation;
the iteration module is used for calculating a loss function of the detection model according to the first sidewalk coordinate, updating and iterating the detection model until the detection model is converged, and obtaining a trained detection model; and
and the obtaining module is used for receiving the original image, inputting the original image into the trained detection model, obtaining a second sidewalk coordinate, and mapping the second sidewalk coordinate to a coordinate on the original image.
9. A computer device comprising a memory having computer readable instructions stored therein and a processor that when executed performs the steps of the pedestrian detection method of any one of claims 1 to 7.
10. A computer-readable storage medium, having computer-readable instructions stored thereon, which, when executed by a processor, implement the steps of the sidewalk detection method according to any one of claims 1 to 7.
CN202010923999.2A 2020-09-04 2020-09-04 Pavement detection method, device, computer equipment and storage medium Active CN112016503B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010923999.2A CN112016503B (en) 2020-09-04 2020-09-04 Pavement detection method, device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010923999.2A CN112016503B (en) 2020-09-04 2020-09-04 Pavement detection method, device, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112016503A true CN112016503A (en) 2020-12-01
CN112016503B CN112016503B (en) 2024-01-23

Family

ID=73516531

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010923999.2A Active CN112016503B (en) 2020-09-04 2020-09-04 Pavement detection method, device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112016503B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112633127A (en) * 2020-12-18 2021-04-09 深圳赛安特技术服务有限公司 Bus lane detection method, device and medium based on image recognition

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108596090A (en) * 2018-04-24 2018-09-28 北京达佳互联信息技术有限公司 Facial image critical point detection method, apparatus, computer equipment and storage medium
WO2019085793A1 (en) * 2017-11-01 2019-05-09 腾讯科技（深圳）有限公司 Image classification method, computer device and computer readable storage medium
CN109815886A (en) * 2019-01-21 2019-05-28 南京邮电大学 A kind of pedestrian and vehicle checking method and system based on improvement YOLOv3
CN109934121A (en) * 2019-02-21 2019-06-25 江苏大学 A kind of orchard pedestrian detection method based on YOLOv3 algorithm
CN110348376A (en) * 2019-07-09 2019-10-18 华南理工大学 A kind of pedestrian's real-time detection method neural network based
CN111062413A (en) * 2019-11-08 2020-04-24 深兰科技（上海）有限公司 Road target detection method and device, electronic equipment and storage medium
CN111126399A (en) * 2019-12-28 2020-05-08 苏州科达科技股份有限公司 Image detection method, device and equipment and readable storage medium

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019085793A1 (en) * 2017-11-01 2019-05-09 腾讯科技（深圳）有限公司 Image classification method, computer device and computer readable storage medium
CN108596090A (en) * 2018-04-24 2018-09-28 北京达佳互联信息技术有限公司 Facial image critical point detection method, apparatus, computer equipment and storage medium
CN109815886A (en) * 2019-01-21 2019-05-28 南京邮电大学 A kind of pedestrian and vehicle checking method and system based on improvement YOLOv3
CN109934121A (en) * 2019-02-21 2019-06-25 江苏大学 A kind of orchard pedestrian detection method based on YOLOv3 algorithm
CN110348376A (en) * 2019-07-09 2019-10-18 华南理工大学 A kind of pedestrian's real-time detection method neural network based
CN111062413A (en) * 2019-11-08 2020-04-24 深兰科技（上海）有限公司 Road target detection method and device, electronic equipment and storage medium
CN111126399A (en) * 2019-12-28 2020-05-08 苏州科达科技股份有限公司 Image detection method, device and equipment and readable storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李大华;汪宏威;高强;于晓;沈洪宇;: "一种卷积神经网络的车辆和行人检测算法", 激光杂志, no. 04 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112633127A (en) * 2020-12-18 2021-04-09 深圳赛安特技术服务有限公司 Bus lane detection method, device and medium based on image recognition

Also Published As

Publication number Publication date
CN112016503B (en) 2024-01-23

Similar Documents

Publication Publication Date Title
WO2022126970A1 (en) Method and device for financial fraud risk identification, computer device, and storage medium
CN112116155A (en) Population mobility prediction method and device based on intelligent decision and computer equipment
CN112329659A (en) Weak supervision semantic segmentation method based on vehicle image and related equipment thereof
CN110163188B (en) Video processing and method, device and equipment for embedding target object in video
US11887473B2 (en) Road congestion detection method and device, and electronic device
WO2023035531A1 (en) Super-resolution reconstruction method for text image and related device thereof
EP3916634A2 (en) Text recognition method and device, and electronic device
CN112016502B (en) Safety belt detection method, safety belt detection device, computer equipment and storage medium
CN112686243A (en) Method and device for intelligently identifying picture characters, computer equipment and storage medium
JP2023527615A (en) Target object detection model training method, target object detection method, device, electronic device, storage medium and computer program
US20230104225A1 (en) Method for fusing road data to generate a map, electronic device, and storage medium
CN112991274A (en) Crowd counting method and device, computer equipment and storage medium
CN115760886B (en) Land parcel dividing method and device based on unmanned aerial vehicle aerial view and related equipment
US20230102804A1 (en) Method of rectifying text image, training method, electronic device, and medium
US20240037911A1 (en) Image classification method, electronic device, and storage medium
CN112651399B (en) Method for detecting same-line characters in inclined image and related equipment thereof
EP3869398A2 (en) Method and apparatus for processing image, device and storage medium
CN112016503B (en) Pavement detection method, device, computer equipment and storage medium
CN114724113B (en) Road sign recognition method, automatic driving method, device and equipment
CN113139490B (en) Image feature matching method and device, computer equipment and storage medium
CN112395450A (en) Picture character detection method and device, computer equipment and storage medium
CN114764890A (en) Pedestrian passageway environment assessment method and device and electronic equipment
CN114764858B (en) Copy-paste image identification method and device, computer equipment and storage medium
CN112052773B (en) Unmanned aerial vehicle traffic dispersion intelligent broadcasting method and device based on image sensing
CN114638810B (en) Document quality analysis method, device, computer 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