CN112818760A - Method for intelligently evaluating influence of weeds on road surface traffic - Google Patents
Method for intelligently evaluating influence of weeds on road surface traffic Download PDFInfo
- Publication number
- CN112818760A CN112818760A CN202110050308.7A CN202110050308A CN112818760A CN 112818760 A CN112818760 A CN 112818760A CN 202110050308 A CN202110050308 A CN 202110050308A CN 112818760 A CN112818760 A CN 112818760A
- Authority
- CN
- China
- Prior art keywords
- weeds
- road surface
- feature
- image
- characteristic
- 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.)
- Withdrawn
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
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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]
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Abstract
The application relates to intelligent state monitoring in the field of intelligent transportation, and particularly discloses a method for intelligently evaluating the influence of weeds on road passing, which is used for carrying out feature recognition on road images based on a computer vision technology and converting the problem of evaluating the passing ability into a classification problem based on recognized features. In particular, in the evaluation process, the method adopts the idea of random walk by a convolution depth neural network to express the extending characteristic of the weeds extending from the gaps outwards in a high-dimensional space by determining the pixel number of the farthest position of the weeds extending in the high-dimensional characteristic of the road surface image from the characteristic position of the gaps, and then, carrying out one-dimensional convolution processing on the characteristic vectors for representing the extending characteristic of the weeds and passing through a classifier to obtain a classification result for representing whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard, so that the classification accuracy is improved.
Description
Technical Field
The present invention relates to intelligent state monitoring in the field of intelligent traffic, and more particularly, to a method for intelligently evaluating the influence of weeds on road surface traffic, a system for intelligently evaluating the influence of weeds on road surface traffic, and an electronic device.
Background
At present, cement ground is generally adopted on low-grade road surfaces in villages, gaps can be reserved on the cement ground after long-time wind, rain and sun exposure, and due to the fact that various weeds exist in the field environment, weeds can grow in the gaps after a period of time, and road surface passing is affected. Because the existing cement ground gap weeding needs to be manually removed by hiring workers, although the hiring cost is low, the efficiency is low, and meanwhile, the weeds shoveled from the cement ground gap need to be cleaned by the workers, so that the labor force is further increased.
Therefore, an optimized solution for intelligently evaluating the influence of weeds on road traffic is desired.
At present, deep learning and neural networks have been widely applied in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
Deep learning and development of a neural network provide a new solution for intelligently evaluating the influence of weeds on road traffic.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. Embodiments of the present application provide a method for intelligently evaluating the influence of weeds on road surface traffic, a system and an electronic device for intelligently evaluating the influence of weeds on road surface traffic by feature recognition of road surface images based on computer vision techniques and converting the above problem of evaluating traffic capacity into a classification problem based on the recognized features. In particular, in the evaluation process, the method adopts the idea of random walk by a convolution depth neural network to express the extending characteristic of the weeds extending from the gaps outwards in a high-dimensional space by determining the pixel number of the farthest position of the weeds extending in the high-dimensional characteristic of the road surface image from the characteristic position of the gaps, and then, carrying out one-dimensional convolution processing on the characteristic vectors for representing the extending characteristic of the weeds and passing through a classifier to obtain a classification result for representing whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard, so that the classification accuracy is improved.
According to one aspect of the present application, there is provided a method for intelligently evaluating the effect of weeds on road traffic, comprising:
acquiring an image to be detected, wherein the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface;
passing the image to be detected through a depth convolution neural network to extract a characteristic map from the image to be detected;
determining a linear region of interest corresponding to a gap of the road surface in the characteristic diagram to obtain a group of characteristic positions;
determining weed objects growing in gaps of the road surface in the feature map, and obtaining the farthest position of the extension of the weed objects corresponding to each feature position to obtain a feature vector formed by the pixel number of a group of pixel points at the farthest position from the feature position;
performing one-dimensional convolution processing on the feature vectors to extract classified feature vectors from the feature vectors; and
and passing the classified characteristic vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not.
In the above method for intelligently evaluating the influence of weeds on road surface traffic, determining a linear region of interest corresponding to a gap of the road surface in the feature map to obtain a set of feature positions, including: identifying a gap object of the road surface in the characteristic diagram by an image semantic segmentation technology; and determining the positions of the pixel points of the gap object as the group of characteristic positions.
In the above method for intelligently evaluating the influence of weeds on road passage, determining the weed objects in the feature map that grow in the gaps of the road surface comprises: and identifying weed objects growing in gaps of the road surface in the feature map by an image semantic segmentation technology.
In the above method for intelligently evaluating the influence of weeds on road traffic, the classifying feature vector is passed through a classifier to obtain a classification result, including: passing the classification feature vector through a Softmax classification function to obtain a first probability that weeds growing on the road surface have an effect on road surface trafficability exceeding a predetermined criterion and a second probability that weeds growing on the road surface have no effect on road surface trafficability exceeding the predetermined criterion; and determining whether the influence of the weeds growing on the road surface passing capacity exceeds a predetermined standard based on the first probability and the second probability. .
In the above method for intelligently evaluating the influence of weeds on road traffic, the deep convolutional neural network is a deep residual network.
According to another aspect of the present application, a system for intelligently evaluating the effects of weeds on road traffic, comprising:
the device comprises an image acquisition unit to be detected, a data acquisition unit and a data acquisition unit, wherein the image acquisition unit to be detected is used for acquiring an image to be detected, the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface;
the characteristic diagram generating unit is used for enabling the image to be detected obtained by the image to be detected obtaining unit to pass through a depth convolution neural network so as to extract a characteristic diagram from the image to be detected;
the characteristic position determining unit is used for determining a linear region of interest corresponding to the gap of the road surface in the characteristic diagram obtained by the characteristic diagram generating unit so as to obtain a group of characteristic positions;
a feature vector generating unit configured to determine a weed object growing in a gap of the road surface in the feature map obtained by the feature map generating unit, and obtain a farthest position at which the weed object extends corresponding to each feature position, so as to obtain a feature vector formed by a group of pixel points at the farthest position from the number of pixels at the feature position obtained by the feature position determining unit;
a classification feature vector generation unit configured to perform one-dimensional convolution processing on the feature vectors obtained by the feature vector generation unit to extract classification feature vectors from the feature vectors; and
and the classification result generating unit is used for enabling the classification feature vector obtained by the classification feature vector generating unit to pass through a classifier so as to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not.
In the above system for intelligently evaluating an influence of weeds on road passage, the characteristic position determination unit includes: the object identification subunit is used for identifying the gap object of the road surface in the feature map by an image semantic segmentation technology; and the determining subunit is used for determining the pixel point position of the gap object obtained by the object identification subunit as the group of characteristic positions.
In the above system for intelligently evaluating an influence of weeds on road passage, the feature vector generation unit is further configured to: and identifying weed objects growing in gaps of the road surface in the feature map by an image semantic segmentation technology.
In the above system for intelligently evaluating an influence of weeds on road passage, the classification result generation unit includes: a probability generating subunit, configured to pass the classification feature vector through a Softmax classification function to obtain a first probability that the influence of the weeds growing on the road surface exceeds a predetermined standard and a second probability that the influence of the weeds growing on the road surface does not exceed the predetermined standard; and a classification result determination subunit operable to determine whether an influence of weeds growing on the road surface on road surface passing ability exceeds a predetermined criterion, based on the first probability and the second probability obtained by the probability generation subunit.
In the above system for intelligently evaluating the influence of weeds on road traffic, the deep convolutional neural network is a deep residual network.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory having stored therein computer program instructions which, when executed by the processor, cause the processor to perform a method for intelligently assessing the effect of weeds on road traffic as described above.
According to yet another aspect of the present application, a computer-readable medium is provided, having stored thereon computer program instructions, which, when executed by a processor, cause the processor to perform the method for intelligently assessing the effect of weeds on road passage as described above.
Compared with the prior art, the method for intelligently evaluating the influence of the weeds on the road surface traffic, the system for intelligently evaluating the influence of the weeds on the road surface traffic and the electronic equipment perform feature recognition on road surface images based on a computer vision technology, and convert the problem of evaluating traffic capacity into a classification problem based on the recognized features. In particular, in the evaluation process, the method adopts the idea of random walk by a convolution depth neural network to express the extending characteristic of the weeds extending from the gaps outwards in a high-dimensional space by determining the pixel number of the farthest position of the weeds extending in the high-dimensional characteristic of the road surface image from the characteristic position of the gaps, and then, carrying out one-dimensional convolution processing on the characteristic vectors for representing the extending characteristic of the weeds and passing through a classifier to obtain a classification result for representing whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard, so that the classification accuracy is improved.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 illustrates an application scenario diagram of a method for intelligently evaluating the effect of weeds on road traffic according to an embodiment of the application;
FIG. 2 illustrates a flow chart of a method for intelligently assessing the impact of weeds on road traffic in accordance with an embodiment of the present application;
FIG. 3 illustrates a system architecture diagram of a method for intelligently evaluating the impact of weeds on road traffic, in accordance with an embodiment of the present application;
FIG. 4 illustrates a flow chart for determining linear regions of interest corresponding to gaps on the road surface in the feature map to obtain a set of feature locations in a method for intelligently evaluating the effect of weeds on road surface traffic according to an embodiment of the present application;
FIG. 5 illustrates a flow chart of passing the classification feature vectors through a classifier to obtain classification results in a method for intelligently evaluating the effect of weeds on road traffic according to an embodiment of the present application;
FIG. 6 illustrates a block diagram of a system for intelligently evaluating the effects of weeds on road traffic, according to an embodiment of the present application.
Fig. 7 illustrates a block diagram of a characteristic position determination unit in a system for intelligently evaluating the effect of weeds on road traffic according to an embodiment of the application.
Fig. 8 illustrates a block diagram of a classification result generation unit in a system for intelligently evaluating the influence of weeds on road traffic according to an embodiment of the present application.
FIG. 9 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Overview of a scene
As mentioned above, the current weeding at the gaps of the cement ground needs to be manually removed by hiring workers, the hiring cost is low, but the efficiency is low, and meanwhile, the weeds shoveled out of the gaps of the cement ground need to be cleaned by the workers, so that the labor force is further increased.
Based on this, the present application is expected to provide a method for evaluating the influence of weeds growing on a cement ground on the road surface trafficability, so that weeding work is performed only when the influence reaches a certain degree, thereby improving the utilization efficiency of labor force.
Here, the influence of weeds on the trafficability of the road surface is mainly determined by the property of weeds extending outward from the gap, and therefore, the inventors of the present application expect to perform feature recognition on the road surface image by computer vision technology and convert the above problem of evaluating trafficability into a classification problem based on the recognized features.
As described above, since the influence of the weeds on the road surface is mainly determined by the property that the weeds extend outward from the gaps, in order to express the property that a specific object in an image extends outward from a certain position, a similar idea to the skeleton generation based on random walk in the image is adopted in the scheme of the present application. That is, after an image of the cement road surface is obtained and input to the convolutional neural network to obtain a feature map, a corresponding linearly arranged set of feature positions, for example, n feature positions, in the feature map is first located by detection of the gap, then, by identification of weeds extending from the gap, the farthest position from which the weeds extend corresponding to each feature position is determined, and the number of pixels of the farthest position from the feature position is obtained, thereby obtaining a feature vector of which the dimension is n. That is, for the feature vector, it expresses not a feature value of a specific position in the feature map but a pixel distance between feature values having a predetermined relationship in the feature map, thereby expressing an extended feature of a predetermined object from a predetermined position in the high-dimensional space.
Then, based on the feature vector, the mutual relation among the feature values of the positions of the original image is further extracted through a plurality of one-dimensional convolutions, so that the distribution features of the original image, extending outwards from the weeds, in the extending dimension of the whole gap are further expressed, and the classification feature vector is obtained.
Finally, the classified feature vector is input into a classifier to obtain a classification result, and the classification result is used for indicating whether the influence of the weeds growing on the cement ground on the road surface passing capacity exceeds a preset standard or not.
Based on this, the present application proposes a method for intelligently assessing the effect of weeds on road traffic, comprising: acquiring an image to be detected, wherein the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface; passing the image to be detected through a depth convolution neural network to extract a characteristic map from the image to be detected; determining a linear region of interest corresponding to a gap of the road surface in the characteristic diagram to obtain a group of characteristic positions; determining weed objects growing in gaps of the road surface in the feature map, and obtaining the farthest position of the extension of the weed objects corresponding to each feature position to obtain a feature vector formed by the pixel number of a group of pixel points at the farthest position from the feature position; performing one-dimensional convolution processing on the feature vectors to extract classified feature vectors from the feature vectors; and passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not.
Fig. 1 illustrates an application scenario diagram of a method for intelligently evaluating the influence of weeds on road traffic according to an embodiment of the application.
As shown in fig. 1, in this application scenario, an image of a road surface to be detected containing weeds is acquired by a camera (e.g., C as illustrated in fig. 1); the image is then input into a server (e.g., S as illustrated in fig. 1) deployed with an algorithm for intelligently assessing the effect of weeds on road traffic, wherein the server is capable of processing the image with the algorithm for intelligently assessing the effect of weeds on road traffic to generate a detection result of whether the effect of weeds growing on the road surface on road traffic capacity exceeds a predetermined criterion.
Having described the general principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
Exemplary method
Fig. 2 illustrates a flow chart of a method for intelligently assessing the effect of weeds on road traffic. As shown in fig. 2, a method for intelligently evaluating the influence of weeds on road traffic according to an embodiment of the present application includes: s110, acquiring an image to be detected, wherein the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface; s120, the image to be detected passes through a depth convolution neural network to extract a characteristic map from the image to be detected; s130, determining a linear region of interest corresponding to the gap of the road surface in the characteristic diagram to obtain a group of characteristic positions; s140, determining weed objects growing in gaps of the road surface in the characteristic diagram, and obtaining the farthest position of the extension of the weed objects corresponding to each characteristic position to obtain a characteristic vector formed by a group of pixel points at the farthest position and the number of pixels at the characteristic position; s150, performing one-dimensional convolution processing on the feature vectors to extract classified feature vectors from the feature vectors; and S160, passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not.
Fig. 3 illustrates an architectural schematic of a method for intelligently assessing the effect of weeds on road traffic according to an embodiment of the application. As shown IN fig. 3, IN the network architecture of the method for intelligently evaluating the influence of weeds on road surface traffic, first, an image to be detected (for example, IN1 as illustrated IN fig. 3) of a road surface containing weeds acquired by a camera is input to a deep convolutional neural network (for example, CNN as illustrated IN fig. 3) to obtain a feature map (for example, F1 as illustrated IN fig. 3); then, determining a linear region of interest corresponding to a gap of the road surface in the feature map to obtain a set of feature positions (e.g., as illustrated in fig. 3 as a 1); then, determining weed objects growing in gaps of the road surface in the feature map, and obtaining a farthest position where the weed objects corresponding to each feature position extend, to obtain a feature vector (for example, as illustrated in fig. 3, V1) composed of a set of pixel points of the farthest position from the number of pixels of the feature position; then, performing one-dimensional convolution processing on the feature vectors to extract classification feature vectors (for example, Vc as illustrated in fig. 3) from the feature vectors; the classification feature vector is then passed through a classifier (e.g., circle S as illustrated in fig. 3) to obtain a classification result, wherein the classification result is used to indicate whether the influence of weeds growing on the road surface throughput exceeds a predetermined criterion.
In step S110, an image to be detected, which is an image of a road surface containing weeds growing in gaps of the road surface, is acquired. As mentioned above, since the cement ground is generally adopted for the low-grade road surface in the country, gaps can be left on the cement ground after long-time wind, rain and sun exposure, and various weeds exist in the field environment, so that the weeds can grow out from the gaps after a period of time, and the road surface passing is influenced. Specifically, in the present embodiment, an image of a road surface containing weeds is first captured by a camera.
In step S120, the image to be detected is passed through a depth convolution neural network to extract a feature map from the image to be detected. Namely, extracting each high-dimensional feature in the image to be detected by using a deep convolutional neural network.
In particular, in embodiments of the present application, the deep convolutional neural network may employ a deep residual neural network, e.g., ResNet 50. It should be known to those skilled in the art that, compared to the conventional convolutional neural network, the deep residual network is an optimized network structure proposed on the basis of the conventional convolutional neural network, which mainly solves the problem of gradient disappearance during the training process. The depth residual error network introduces a residual error network structure, the network layer can be made deeper through the residual error network structure, and the problem of gradient disappearance can not occur. The residual error network uses the cross-layer link thought of a high-speed network for reference, breaks through the convention that the traditional neural network only can provide N layers as input from the input layer of the N-1 layer, enables the output of a certain layer to directly cross several layers as the input of the later layer, and has the significance of providing a new direction for the difficult problem that the error rate of the whole learning model is not reduced and inversely increased by superposing multiple layers of networks.
In step S130, a linear region of interest corresponding to a gap of the road surface in the feature map is determined to obtain a set of feature positions. That is, a corresponding linearly arranged set of feature positions in the feature map is located through the detection of the gap.
Specifically, in the embodiment of the present application, the process of determining a linear region of interest corresponding to a gap of the road surface in the feature map to obtain a set of feature positions includes: first, the slit objects of the road surface in the feature map are identified by an image semantic segmentation technique, and those skilled in the art should know that the image semantic segmentation technique can classify the semantics of each region (i.e. what object the region is). Then, the positions of the pixel points of the gap object are determined as the set of characteristic positions.
It should be noted that in other examples of the present application, the gap object may also be determined in other manners, for example, a rectangular candidate frame corresponding to the gap of the road surface is selected from the feature map in a manner of a target candidate frame, and a most direct manner of extracting a candidate region is to use a sliding window method; or, through a target candidate frame extraction network, such as the Faster R-CNN, the most core work of the Faster R-CNN is to complete the candidate frame extraction operation in the process of extracting the characteristics, thereby greatly accelerating the speed of object detection; still alternatively, an abnormal feature, i.e., a gap object, in the image is determined by an image abnormal pixel detection method, where the method of determining the gap object is not limited by the present application.
Fig. 4 illustrates a flowchart for determining a linear region of interest corresponding to a gap of the road surface in the feature map to obtain a set of feature positions in the method for intelligently evaluating the influence of weeds on road surface traffic according to the embodiment of the application. As shown in fig. 4, determining a linear region of interest corresponding to a gap of the road surface in the feature map to obtain a set of feature positions includes: s210, identifying a gap object of the road surface in the feature map through an image semantic segmentation technology; and S220, determining the positions of the pixel points of the gap object as the group of characteristic positions.
In step S140, weed objects growing in gaps of the road surface in the feature map are determined, and the farthest position where the weed objects extend corresponding to each feature position is obtained, so as to obtain a feature vector formed by a group of pixel points at the farthest position from the number of pixels at the feature position. It should be understood that since the influence of weeds on the road surface is mainly determined by the property of weeds extending outwards from the gaps, in order to express the property of a specific object in an image extending outwards from a certain position, a similar idea to the skeleton generation based on random walk in the image is adopted in the scheme of the application. That is, by identifying the weeds extending from the gap, the farthest position from which the weeds corresponding to each feature position extend is determined, and the number of pixels from which the farthest position is distant from the feature position is obtained, thereby obtaining the feature vector whose dimension is n. That is, for the feature vector, it expresses not a feature value of a specific position in the feature map but a pixel distance between feature values having a predetermined relationship in the feature map, thereby expressing an extended feature of a predetermined object from a predetermined position in the high-dimensional space.
Specifically, in the embodiment of the present application, the process of determining the weed objects growing in the gaps of the road surface in the feature map includes: and identifying weed objects growing in gaps of the road surface in the feature map by an image semantic segmentation technology. One of ordinary skill in the art will appreciate that image semantic segmentation techniques can classify the semantics of each block region. It should be noted that in other examples of the present application, the gap object may also be determined in other manners, for example, in a manner of a target candidate frame, by extracting a network from the target candidate frame, or by detecting an abnormal pixel in an image, and the like, and herein is not limited by the present application.
In step S150, a one-dimensional convolution process is performed on the feature vectors to extract classified feature vectors from the feature vectors. That is, the correlation among the feature values of the positions of the original image is extracted through a one-dimensional convolution neural network, so that the distribution features of the original image extending outwards of the weeds in the extending dimension of the whole gap are further expressed, and a classification feature vector is obtained.
In step S160, the classified feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a predetermined standard.
Specifically, in the embodiment of the present application, the process of passing the classification feature vector through a classifier to obtain a classification result includes: first, the classification feature vector is passed through a Softmax classification function to obtain a first probability that the influence of the weeds growing on the road surface exceeds a predetermined criterion and a second probability that the influence of the weeds growing on the road surface does not exceed the predetermined criterion, the Softmax classification function being a normalized exponential function that is never satisfactory for the score, as will be appreciated by those skilled in the art: a correct classification can always result in a higher probability, an incorrect classification can always result in a lower probability, and the loss value can always be smaller. Then, based on the first probability and the second probability, whether the influence of the weeds growing on the road surface passing capacity exceeds a predetermined standard is determined, namely, the larger of the first probability and the second probability is determined as a final classification result.
Fig. 5 illustrates a flowchart for intelligently evaluating the influence of weeds on road traffic in a method for intelligently evaluating the influence of weeds on road traffic, according to an embodiment of the present application, and passing the classification feature vectors through a classifier to obtain a classification result. As shown in fig. 5, passing the classified feature vector through a classifier to obtain a classification result, including: s310, passing the classification feature vector through a Softmax classification function to obtain a first probability that the influence of the weeds growing on the road surface passing capacity exceeds a preset standard and a second probability that the influence of the weeds growing on the road surface not exceeds the preset standard; and S320, determining whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not based on the first probability and the second probability.
In summary, the method for intelligently evaluating the influence of weeds on road traffic is illustrated, which is based on computer vision technology to perform feature recognition on road images and converts the above problem of evaluating traffic capacity into a classification problem based on the recognized features. Particularly, the extending characteristic of the weeds in the high-dimensional space extending outwards from the gaps is expressed by determining the pixel number of the farthest position of the extension of the weeds in the high-dimensional characteristic of the road surface image from the characteristic position of the gaps, and then the distribution characteristic of the extension of the weeds in the original image in the whole extending dimension of the gaps is further extracted through a plurality of one-dimensional convolutions, so that the classification accuracy is improved.
Exemplary System
FIG. 6 illustrates a block diagram of a system for intelligently evaluating the effects of weeds on road traffic, according to an embodiment of the present application.
As shown in fig. 6, a system 600 for intelligently evaluating the effect of weeds on road traffic according to an embodiment of the present application includes: the image acquiring unit 610 to be detected is used for acquiring an image to be detected, wherein the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface; a feature map generation unit 620, configured to pass the to-be-detected image obtained by the to-be-detected image obtaining unit 610 through a depth convolution neural network, so as to extract a feature map from the to-be-detected image; a characteristic position determining unit 630, configured to determine a linear region of interest corresponding to a gap of the road surface in the characteristic map obtained by the characteristic map generating unit 620, so as to obtain a set of characteristic positions; a feature vector generating unit 640 configured to determine weed objects growing in gaps of the road surface in the feature map obtained by the feature map generating unit 620, and obtain the farthest position where the weed objects extend corresponding to each feature position, so as to obtain a feature vector formed by a group of pixels at the farthest position from the number of pixels at the feature position obtained by the feature position determining unit 630; a classification feature vector generation unit 650 configured to perform one-dimensional convolution processing on the feature vectors obtained by the feature vector generation unit 640 to extract classification feature vectors from the feature vectors; and a classification result generating unit 660, configured to pass the classification feature vector obtained by the classification feature vector generating unit 650 through a classifier to obtain a classification result, where the classification result is used to indicate whether the influence of the weeds growing on the road surface passing capacity exceeds a predetermined standard.
In one example, in the above system 600 for intelligently evaluating the influence of weeds on road traffic, as shown in fig. 7, the characteristic position determination unit 630 includes: an object identification subunit 631, configured to identify a gap object of the road surface in the feature map by using an image semantic segmentation technique; and a determining subunit 632 configured to determine the pixel point positions of the slit object obtained by the object identifying subunit 631 as the set of feature positions.
In one example, in the system 600 for intelligently evaluating the influence of weeds on road traffic described above, the feature vector generating unit 640 is further configured to: and identifying weed objects growing in gaps of the road surface in the feature map by an image semantic segmentation technology.
In one example, in the system 600 for intelligently evaluating the influence of weeds on road traffic as described above, as shown in fig. 8, the classification result generating unit 660 includes: a probability generating sub-unit 661, configured to pass the classification feature vector through a Softmax classification function to obtain a first probability that the influence of the weeds growing on the road surface exceeds a predetermined criterion and a second probability that the influence of the weeds growing on the road surface does not exceed the predetermined criterion; and a classification result determining subunit 662 configured to determine whether the influence of the weeds growing on the road surface traffic capacity exceeds a predetermined criterion, based on the first probability and the second probability obtained by the probability generating subunit 661.
In one example, in the system 600 for intelligently evaluating the impact of weeds on road traffic described above, the deep convolutional neural network is a deep residual network.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described intelligent evaluation system 600 have been described in detail in the above description of the method for intelligently evaluating the influence of weeds on road surface traffic with reference to fig. 1 to 5, and therefore, a repetitive description thereof will be omitted.
As described above, the intelligent evaluation system 600 according to the embodiment of the present application can be implemented in various terminal devices, such as a server for evaluating the influence of weeds on road passage, and the like. In one example, the intelligent evaluation system 600 according to the embodiment of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the intelligent evaluation system 600 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the intelligent evaluation system 600 may also be one of many hardware modules of the terminal device.
Alternatively, in another example, the intelligent evaluation system 600 and the terminal device may be separate devices, and the intelligent evaluation system 600 may be connected to the terminal device through a wired and/or wireless network and transmit the mutual information according to an agreed data format.
Exemplary electronic device
Next, an electronic apparatus according to an embodiment of the present application is described with reference to fig. 9.
FIG. 9 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.
As shown in fig. 9, the electronic device 10 includes one or more processors 11 and a memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
In one example, the electronic device 10 may further include: an input system 13 and an output system 14, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
The input system 13 may comprise, for example, a keyboard, a mouse, etc.
The output system 14 may output various information including classification results and the like to the outside. The output system 14 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, among others.
Of course, for simplicity, only some of the components of the electronic device 10 relevant to the present application are shown in fig. 9, and components such as buses, input/output interfaces, and the like are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the functions in the method for intelligently evaluating the effect of weeds on road passage according to various embodiments of the present application described in the "exemplary methods" section above in this specification.
The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the steps in the method for intelligently evaluating the effect of weeds on road passage described in the "exemplary methods" section above in this specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.
The block diagrams of devices, apparatuses, systems referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (10)
1. A method for intelligently assessing the effects of weeds on road traffic, comprising:
acquiring an image to be detected, wherein the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface;
passing the image to be detected through a depth convolution neural network to extract a characteristic map from the image to be detected;
determining a linear region of interest corresponding to a gap of the road surface in the characteristic diagram to obtain a group of characteristic positions;
determining weed objects growing in gaps of the road surface in the feature map, and obtaining the farthest position of the extension of the weed objects corresponding to each feature position to obtain a feature vector formed by the pixel number of a group of pixel points at the farthest position from the feature position;
performing one-dimensional convolution processing on the feature vectors to extract classified feature vectors from the feature vectors; and
and passing the classified characteristic vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not.
2. The method for intelligently evaluating the impact of weeds on road traffic as claimed in claim 1, wherein determining a linear region of interest corresponding to a gap of the road surface in the feature map to obtain a set of feature locations comprises:
identifying a gap object of the road surface in the characteristic diagram by an image semantic segmentation technology; and
and determining the positions of the pixel points of the gap object as the group of characteristic positions.
3. The method for intelligently evaluating the impact of weeds on road traffic as claimed in claim 1, wherein determining weed objects in the signature that grow in the crevices of the road surface comprises:
and identifying weed objects growing in gaps of the road surface in the feature map by an image semantic segmentation technology.
4. The method for intelligently evaluating the impact of weeds on road traffic as claimed in claim 1, wherein passing the classification feature vector through a classifier to obtain a classification result comprises:
passing the classification feature vector through a Softmax classification function to obtain a first probability that weeds growing on the road surface have an effect on road surface trafficability exceeding a predetermined criterion and a second probability that weeds growing on the road surface have no effect on road surface trafficability exceeding the predetermined criterion; and
determining whether the influence of the weeds growing on the road surface passing capacity exceeds a predetermined criterion based on the first probability and the second probability.
5. The method for intelligently evaluating the impact of weeds on road traffic as claimed in claim 1, wherein said deep convolutional neural network is a deep residual network.
6. A system for intelligently evaluating the effects of weeds on road traffic, comprising:
the device comprises an image acquisition unit to be detected, a data acquisition unit and a data acquisition unit, wherein the image acquisition unit to be detected is used for acquiring an image to be detected, the image to be detected is an image of a road surface containing weeds, and the weeds grow in gaps of the road surface;
the characteristic diagram generating unit is used for enabling the image to be detected obtained by the image to be detected obtaining unit to pass through a depth convolution neural network so as to extract a characteristic diagram from the image to be detected;
the characteristic position determining unit is used for determining a linear region of interest corresponding to the gap of the road surface in the characteristic diagram obtained by the characteristic diagram generating unit so as to obtain a group of characteristic positions;
a feature vector generating unit configured to determine a weed object growing in a gap of the road surface in the feature map obtained by the feature map generating unit, and obtain a farthest position at which the weed object extends corresponding to each feature position, so as to obtain a feature vector formed by a group of pixel points at the farthest position from the number of pixels at the feature position obtained by the feature position determining unit;
a classification feature vector generation unit configured to perform one-dimensional convolution processing on the feature vectors obtained by the feature vector generation unit to extract classification feature vectors from the feature vectors; and
and the classification result generating unit is used for enabling the classification feature vector obtained by the classification feature vector generating unit to pass through a classifier so as to obtain a classification result, wherein the classification result is used for indicating whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not.
7. The system for intelligently evaluating the impact of weeds on road traffic according to claim 6, wherein said characteristic location determination unit comprises:
the object identification subunit is used for identifying the gap object of the road surface in the feature map by an image semantic segmentation technology; and
and the determining subunit is used for determining the pixel point position of the gap object obtained by the object identification subunit as the group of characteristic positions.
8. The system for intelligently evaluating the influence of weeds on road traffic according to claim 6, wherein said classification result generation unit comprises:
a probability generating subunit, configured to pass the classification feature vector through a Softmax classification function to obtain a first probability that the influence of the weeds growing on the road surface exceeds a predetermined standard and a second probability that the influence of the weeds growing on the road surface does not exceed the predetermined standard; and
and the classification result determining subunit is used for determining whether the influence of the weeds growing on the road surface passing capacity exceeds a preset standard or not based on the first probability and the second probability obtained by the probability generating subunit.
9. The system for intelligently evaluating the impact of weeds on road traffic according to claim 6, wherein said deep convolutional neural network is a deep residual network.
10. An electronic device, comprising:
a processor; and
memory in which computer program instructions are stored which, when executed by the processor, cause the processor to carry out the method for intelligent assessment of the effect of weeds on road passage according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110050308.7A CN112818760A (en) | 2021-01-14 | 2021-01-14 | Method for intelligently evaluating influence of weeds on road surface traffic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110050308.7A CN112818760A (en) | 2021-01-14 | 2021-01-14 | Method for intelligently evaluating influence of weeds on road surface traffic |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112818760A true CN112818760A (en) | 2021-05-18 |
Family
ID=75869539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110050308.7A Withdrawn CN112818760A (en) | 2021-01-14 | 2021-01-14 | Method for intelligently evaluating influence of weeds on road surface traffic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112818760A (en) |
-
2021
- 2021-01-14 CN CN202110050308.7A patent/CN112818760A/en not_active Withdrawn
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112380952B (en) | Power equipment infrared image real-time detection and identification method based on artificial intelligence | |
| CN109978893B (en) | Training method, device, equipment and storage medium of image semantic segmentation network | |
| CN110853033B (en) | Video detection method and device based on inter-frame similarity | |
| CN109977895B (en) | Wild animal video target detection method based on multi-feature map fusion | |
| US11915500B2 (en) | Neural network based scene text recognition | |
| CN110348475A (en) | It is a kind of based on spatial alternation to resisting sample Enhancement Method and model | |
| CN113869449A (en) | Model training method, image processing method, device, equipment and storage medium | |
| CN112560827B (en) | Model training method, model training device, model prediction method, electronic device, and medium | |
| CN115471216B (en) | Data management method of intelligent laboratory management platform | |
| CN111046971A (en) | Image recognition method, device, equipment and computer readable storage medium | |
| CN116311214B (en) | License plate recognition method and device | |
| CN110610123A (en) | Multi-target vehicle detection method and device, electronic equipment and storage medium | |
| CN116089648B (en) | File management system and method based on artificial intelligence | |
| CN110956615A (en) | Image quality evaluation model training method and device, electronic equipment and storage medium | |
| KR101268520B1 (en) | The apparatus and method for recognizing image | |
| US9177215B2 (en) | Sparse representation for dynamic sensor networks | |
| CN111783812A (en) | Method and device for identifying forbidden images and computer readable storage medium | |
| CN112733858B (en) | Image character rapid identification method and device based on character region detection | |
| CN112465805A (en) | Neural network training method for quality detection of steel bar stamping and bending | |
| CN113255557A (en) | Video crowd emotion analysis method and system based on deep learning | |
| CN113255752A (en) | Solid material consistency sorting method based on feature clustering | |
| EP4332910A1 (en) | Behavior detection method, electronic device, and computer readable storage medium | |
| CN112818760A (en) | Method for intelligently evaluating influence of weeds on road surface traffic | |
| CN112115928B (en) | Training method and detection method of neural network based on illegal parking vehicle labels | |
| Li et al. | Target segmentation of industrial smoke image based on LBP Silhouettes coefficient variant (LBPSCV) algorithm |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210518 |