Adverse road condition is the main cause of traffic accidents. Road surface condition recognition based on video image has become a central issue. However, hybrid road surface and road surface under different lighting environments are two crucial problems. In this paper, the road surface states are categorized into 5 types including dry, wet, snow, ice, and water. Then, according to the original image size, images are segmented; 9-dimensional color eigenvectors and 4 texture eigenvectors are extracted to construct road surface state characteristics database. Next, a recognition method of road surface state based on SVM (Support Vector Machine) is proposed. In order to improve the recognition accuracy and the universality, a grid searching algorithm and PSO (Particle Swarm Optimization) algorithm are used to optimize the kernel function factor and penalty factor of SVM. Finally, a large number of actual road surface images in different environments are tested. The results show that the method based on SVM and image segmentation is feasible. The accuracy of PSO algorithm is more than 90%, which effectively solves the problem of road surface state recognition under the condition of hybrid or different video scenes.
According to statistics, 16.12% of traffic accidents on the highway are ascribed to slippery road conditions [
In the field of traffic meteorology, the road surface state can be categorized in dry, wet, water, snow, and ice types according to different forms of liquid on road surface. At present, the road surface detection sensor is the main entrance to obtain the information of road surface slippery conditions. Cai et al. [
With the widespread application of road surveillance cameras, more and more scholars pay attention to the image processing technology of road surface slippery condition recognition. Andreas and Wilco [
Image feature extraction is a key step in image recognition. Zhang et al. [
Reviewing the above literatures, it is found that the existing problems and development trends of image recognition technology are as follows: Image recognition technology is the main technology of road surface recognition. However, due to the complexity of the road scene and the weak adaptability of the vision system to the illumination change, the road condition detection method based on machine vision has the problem of weak adaptability, the poor robustness of illumination, and low recognition accuracy at present. The identification of the hybrid road surface state is one of the main problems in this study. Using SVM, neural network, and other machine learning methods to identify the road surface state is the development trend. Extracting appropriate multidimensional color and texture eigenvectors can help to improve the accuracy of road surface state recognition.
Therefore, this paper presents a new method based on SVM classifier and image segmentation processing to solve the problem of the small size of the sample and nonlinear and high-dimension pattern recognition. First of all, the comprehensive sample database of road surface state is established by collecting road surface images in different scenes through a variety of ways. Then, 13-dimensional color and texture eigenvectors are extracted to build the training database of road surface state. Next, the optimal parameters of the SVM classifier are trained by the grid searching optimization algorithm and the PSO algorithm, respectively. Thus two kinds of road surface state classification models are built and the performances of the two optimization classification models are compared. For the hybrid road surface state recognition, the road surface state image is segmented into blocks and the overall state of road surface state is presented. Finally, the algorithm proposed is tested and the ideal recognition results are obtained based on the large-scale samples.
The road surface image information mainly includes color, texture, shape, and other characteristics. In this paper, representatively typical road surface state image samples are selected and color and texture eigenvectors are extracted, and the road surface state image feature database can be formed by researching color and the texture characteristics of road state.
Color eigenvectors of road surface image are usually stable and not sensitive to size or direction. Among them, the color moment feature has the characteristics of translation invariance, rotation invariance, and scale invariance, which can ensure the integrity of image color information [
Gray level cooccurrence matrix can better represent the texture information [
Energy reflects the texture thickness of image. When the texture is coarse relatively,
Entropy reflects the amount of the image information. When the image has more textures, the entropy value is larger. If the image contains fewer textures, the entropy value is smaller. If the image has no textures, the entropy value is close to zero.
The contrast reflects the clarity of the image texture. In images, the deeper the texture groove, the greater the contrast, and the clearer the image texture visual effect.
Correlation value reflects the correlation of local gray scale in images. When the values of the matrix elements are evenly equal, the correlation value is large. On the contrary, when the values of the matrix elements are very different, the correlation value is small.
Based on the research above, a set of 13-dimensional road surface state eigenvectors is determined as
As shown in Figure
Road surface images acquisition system.
The basis of road surface state recognition is to establish the road surface state feature database, which needs to collect a large number of road surface state image samples through various ways. Because of the simplicity of the road surface images collected by the experimental system, we also use the highway video surveillance resources, network resources, and other video resources to collect road images to expand the sample database.
The road surface state is divided into five types including dry, wet, water, ice, and snow. According to the influence of original images to samples database under the condition of different images size and lighting scenes, the original image segmentation principle is proposed as shown in Table
Image segmentation principle.
Size of images (px) | Size of blocks (px) |
---|---|
100000 ≤ image < 1000000 |
|
1000000 ≤ image < 2000000 |
|
2000000 ≤ image < 3000000 |
|
3000000 ≤ image < 5000000 |
|
5000000 ≤ image < 8500000 |
|
In this paper, 500 dry images, 500 wet images, 500 water images, 500 snow images, and 500 ice images totaling 2500 images were collected to construct the sample database. Some of the image samples are shown in Figure
Road surface image sample library.
Based on the road surface state image sample database, 500 samples were collected for each state, and the color and texture eigenvectors were extracted to build the road surface state feature database. Figures
Color eigenvector:
Color eigenvector:
Texture eigenvectors: energy.
Texture eigenvectors: entropy.
As shown in Figure
As shown in Figure
As shown in Figure
As shown in Figure
It can be concluded that the single feature curves of the five states have an overlapping area, but there are obvious differences in the feature vectors between at least two kinds of states. The 13-dimensional feature mentioned in this paper can help to accurately identify the road surface state.
The principle of SVM [
Linear SVM classification function is as follows:
For the nonlinear classification function, the existence of misclassified samples is allowed by introducing nonnegative slack variable
In this case, the reciprocal of the maximum classification interval is
After constructing the optimal hyperplane, the most widely used Gaussian kernel function
Then the input vector
Based on the nonlinear optimal classification function, the main idea of multiclassification can be explained as follows: Assuming that a SVM classifier is designed between every two types of samples,
In the process of SVM classification and identification, the penalty factor
Based on the grid searching algorithm, the principle of parameter optimization [
Among them, there will be many combinations corresponding to the highest verification classification accuracy. The combination of the smallest
The basic principle of Particle Swarm Optimization (PSO) [
Parameters optimization based on the particle swarm algorithm is as follows.
Initialize the size and initial velocity of the particle
The fitness value of each particle
The fitness value of each particle
Comparing the fitness value of each particle
According to (
When the end condition is reached, the
Firstly, two SVM parameters optimization algorithms are used to obtain two groups of optimal training parameters
(1) Mark the surface state conditions: dry as D, wet as Wt, water as Wr, snow as S, and ice as I. The eigenvectors of 400 samples of each road state were extracted to form the training database.
(2) The training data is inputted into SVM classifier; the best training parameters
From Table
The road surface condition classification model.
Number of training samples | Parameters optimization time consuming (s) | Optimal |
Training accuracy |
---|---|---|---|
2000 | Grid algorithm 2.6482 | Grid algorithm |
Grid optimization model 90.97% |
PSO algorithm 0.6848 | PSO algorithm |
PSO optimization model 99.12% |
(3) 20% of the sample data were tested by the classification model to verify the recognition performance of the two classification models. The test results are shown in Table
Classification model performance test.
Number of test samples | Test accuracy |
---|---|
500 | Grid optimization model 88.63% |
PSO optimization model 97.02% |
From Table
Firstly, the actual road surface image is divided into blocks according to the segmentation principle. Next, the 13-dimensional feature of each block is extracted. Then the road surface block feature vectors are input into two classification models mentioned above. And the state of each block will be recognized. When all the blocks are recognized, the proportion of each state will be counted.
The recognition results of the dry road surface state under good illumination condition (from the experimental system) are shown in Figure
Recognition results of grid.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | D | D | D | D | D | D | D | D | D |
2 | D | D | D | D | D | D | D | D | D |
3 | D | D | D | D | D | D | D | D | D |
4 | D | D | D | D | D | D | D | D | D |
5 | D | D | D | D | D | D | D | D | D |
Recognition results of PSO.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | D | D | D | D | D | D | D | D | D |
2 | Wt | D | D | D | D | D | D | D | D |
3 | D | D | D | D | Wt | D | D | D | D |
4 | D | D | D | D | D | D | D | D | D |
5 | D | D | D | D | D | D | Wt | D | D |
Image test results of dry road surface state under good lighting condition (from the experimental system). (a) Image of dry road surface. (b) Image blocks.
Table
The dry road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 45 | 42 | 45 | 93.33% | 100% |
Wet | Wt | 3 | 0 | 6.67% | 0 | |
Water | Wr | 0 | 0 | 0 | 0 | |
Snow | S | 0 | 0 | 0 | 0 | |
Ice | I | 0 | 0 | 0 | 0 |
From Table
The recognition results of dry road surface state under adverse lighting conditions (from the experimental system) are shown in Figure
Recognition results of grid.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | D | D | D | D | Wt | D | D | D | D |
2 | Wt | D | D | D | D | D | D | Wt | D |
3 | D | Wt | D | D | Wt | D | D | D | Wt |
4 | D | D | D | Wt | D | D | D | D | Wt |
5 | Wt | D | D | D | D | D | Wt | D | D |
Recognition results of PSO.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | D | D | D | D | D | D | D | D | D |
2 | D | D | D | D | D | D | D | D | D |
3 | D | D | D | D | D | D | D | D | D |
4 | D | D | D | D | Wt | D | D | D | D |
5 | D | D | D | D | D | D | D | Wt | D |
Image test results of dry road surface state under adverse lighting condition (from the experimental system). (a) Image of dry road surface state. (b) Image blocks.
Table
The dry road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 45 | 35 | 43 | 77.78% | 95.56% |
Wet | Wt | 10 | 2 | 22.22% | 4.44% | |
Water | Wr | 0 | 0 | 0 | 0 | |
Snow | S | 0 | 0 | 0 | 0 | |
Ice | I | 0 | 0 | 0 | 0 |
From Table
The recognition results of the wet road surface state under good illumination condition (from the surveillance system) are shown in Figure
Recognition results of grid.
Column | Row | |||||||
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
1 | Wt | Wt | Wt | Wt | Wt | D | Wt | Wt |
2 | Wt | Wt | Wt | Wt | Wt | D | Wt | D |
3 | Wt | Wt | Wt | Wt | Wt | Wt | D | D |
4 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | D |
5 | D | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
6 | D | Wt | Wt | Wt | Wt | Wt | Wt | D |
Recognition results of PSO.
Column | Row | |||||||
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
1 | Wr | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
2 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
3 | Wt | Wt | Wt | Wt | Wt | Wr | Wt | Wt |
4 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
5 | Wt | Wt | Wt | Wr | Wt | Wt | Wt | Wt |
6 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
Image test results of wet road surface state under good lighting condition (from the surveillance system). (a) Image of wet road surface state. (b) Image blocks.
Table
The wet road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 48 | 9 | 0 | 18.75% | 0 |
Wet | Wt | 39 | 45 | 81.25% | 93.75% | |
Water | Wr | 0 | 3 | 0 | 6.25% | |
Snow | S | 0 | 0 | 0 | 0 | |
Ice | I | 0 | 0 | 0 | 0 |
From Table
The recognition results of the wet road surface state under adverse illumination condition (from the experimental system) are shown in Figure
Recognition results of grid.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | Wt | Wt | Wt | Wt | Wt | D | Wt | Wt | D |
2 | Wt | Wt | Wt | Wt | Wt | Wr | Wt | D | Wt |
3 | Wt | Wt | D | Wt | Wt | D | Wt | Wt | Wt |
4 | Wt | Wt | Wt | Wt | Wt | D | Wt | I | Wt |
5 | Wt | Wt | Wt | Wt | Wt | D | Wt | Wt | Wt |
Recognition results of PSO.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wr | Wt |
2 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
3 | Wt | Wt | Wt | I | Wt | Wt | Wt | Wt | Wt |
4 | Wt | Wt | Wt | Wt | I | Wt | Wt | Wt | Wt |
5 | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt | Wt |
Image test results of wet road surface state under good lighting condition (from the experimental system). (a) Image of wet road surface state. (b) Image blocks.
Table
The wet road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 45 | 7 | 0 | 15.56% | 0 |
Wet | Wt | 36 | 42 | 80.00% | 93.33% | |
Water | Wr | 1 | 1 | 2.22% | 2.22% | |
Snow | S | 0 | 0 | 0 | 0 | |
Ice | I | 1 | 2 | 2.22% | 4.44% |
From Table
The recognition results of the water road surface state under good illumination condition (from the mobile camera) are shown in Figure
Recognition results of grid.
Column | Row | ||||||
---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | |
1 | Wr | S | Wr | Wr | Wr | Wr | I |
2 | Wr | Wr | Wr | Wt | Wt | Wr | Wr |
3 | Wr | Wr | Wt | Wr | Wr | I | Wr |
4 | I | Wr | Wr | Wr | Wr | Wr | Wr |
5 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
6 | Wr | Wr | Wr | Wr | Wr | Wr | D |
7 | Wr | S | Wr | Wr | Wr | Wr | I |
8 | I | Wr | Wr | Wr | Wr | Wr | I |
Recognition results of PSO.
Column | Row | ||||||
---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | |
1 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
2 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
3 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
4 | Wr | Wr | Wr | Wr | Wr | Wr | Wt |
5 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
6 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
7 | Wr | Wr | Wr | Wr | Wr | D | Wr |
8 | Wr | Wr | Wr | Wr | Wr | Wr | Wr |
Image test results of water road surface state under good lighting condition (from the mobile camera). (a) Image of water road surface state. (b) Image blocks.
Table
The water road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 56 | 1 | 1 | 1.79% | 1.79% |
Wet | Wt | 3 | 1 | 5.36% | 1.79% | |
Water | Wr | 44 | 54 | 78.57% | 96.42% | |
Snow | S | 2 | 0 | 3.57% | 0 | |
Ice | I | 6 | 0 | 10.71% | 0 |
From Table
The recognition results of the water road surface state with reflection (from the Internet images) are shown in Figure
Recognition results of grid.
Column | Row | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
1 | Wr | S | Wr | Wr | Wr | I |
2 | Wr | Wr | Wr | Wt | Wr | Wr |
3 | Wr | Wt | Wr | Wr | I | Wr |
4 | I | Wr | Wr | D | Wr | Wr |
5 | Wr | Wr | Wr | Wr | Wr | Wr |
6 | Wr | Wr | Wr | Wr | Wr | Wr |
7 | Wr | S | Wr | Wr | Wr | I |
8 | I | Wr | Wr | Wr | Wr | I |
Recognition results of PSO.
Column | Row | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
1 | Wt | Wr | Wr | Wr | Wt | Wr |
2 | Wr | Wr | Wr | Wr | Wt | Wr |
3 | Wr | Wr | Wr | Wr | Wr | Wr |
4 | Wr | Wr | Wr | Wr | Wt | Wr |
5 | Wr | Wr | Wr | Wr | Wr | Wr |
6 | Wr | Wr | Wr | Wr | Wr | Wr |
7 | Wr | Wr | Wr | Wr | Wr | Wr |
8 | I | Wr | Wr | Wr | Wr | Wt |
Image test results of water road surface state with reflection (from the Internet images). (a) Image of water road surface state. (b) Image blocks.
Table
The water road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 48 | 2 | 0 | 4.17% | 0 |
Wet | Wt | 2 | 4 | 4.16% | 8.33% | |
Water | Wr | 36 | 43 | 75% | 89.59% | |
Snow | S | 2 | 0 | 4.17% | 0 | |
Ice | I | 6 | 1 | 12.5% | 2.08% |
From Table
The recognition results of the snow road surface state (from the Internet images) are shown in Figure
Recognition results of grid.
Column | Row | ||||||
---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | |
1 | Wr | S | S | S | S | S | I |
2 | S | I | S | S | S | S | S |
3 | S | I | S | I | S | S | S |
4 | S | I | S | S | I | S | S |
5 | I | S | S | S | S | S | I |
6 | I | I | S | S | S | I | Wr |
7 | I | I | S | S | S | I | S |
Recognition results of PSO.
Column | Row | ||||||
---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | |
1 | Wr | S | S | S | S | S | S |
2 | S | S | S | S | S | S | S |
3 | S | S | S | S | S | S | S |
4 | S | S | S | S | I | S | S |
5 | I | S | S | S | S | S | S |
6 | S | I | S | S | S | I | Wr |
7 | S | S | S | S | S | I | S |
Image test results of snow road surface (from the Internet images). (a) Image of snow road surface state. (b) Image blocks.
Table
The snow road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 49 | 0 | 0 | 0 | 0 |
Wet | Wt | 0 | 0 | 0 | 0 | |
Water | Wr | 2 | 2 | 4.08% | 4.08% | |
Snow | S | 33 | 42 | 67.35% | 85.71% | |
Ice | I | 14 | 5 | 28.57% | 10.20% |
From Table
The recognition results of the snow road surface state (from the surveillance system) are shown in Figure
Recognition results of grid.
Column | Row | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
1 | S | S | S | S | S | I |
2 | D | D | S | S | S | S |
3 | S | S | S | S | D | D |
4 | S | S | S | Wr | D | S |
5 | S | S | S | S | S | S |
6 | S | S | S | S | S | S |
7 | S | S | S | D | Wt | S |
8 | S | S | S | D | Wt | S |
Recognition results of PSO.
Column | Row | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
1 | S | S | S | S | S | I |
2 | S | S | S | S | S | S |
3 | S | S | S | S | S | I |
4 | S | S | S | S | S | S |
5 | S | S | S | S | S | S |
6 | S | S | S | S | S | S |
7 | S | S | S | S | Wt | S |
8 | S | S | S | S | S | I |
Image test results of snow road surface (from the surveillance system). (a) Image of snow road surface state. (b) Image blocks.
Table
The snow road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 49 | 7 | 0 | 14.29% | 0 |
Wet | Wt | 2 | 1 | 4.08% | 2.04% | |
Water | Wr | 1 | 0 | 2.04% | 0% | |
Snow | S | 38 | 45 | 77.55% | 91.84% | |
Ice | I | 1 | 3 | 2.04% | 6.12% |
From Table
The recognition results of the ice road surface state under good illumination condition (from the experimental system) are shown in Figure
Recognition results of grid.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | I | I | I | I | I | I | I | Wt | I |
2 | I | I | I | I | I | I | I | Wt | I |
3 | I | I | I | I | I | I | I | I | I |
4 | I | I | I | I | I | I | I | I | I |
5 | I | I | I | I | I | Wt | I | I | I |
Recognition results of PSO.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | I | I | I | I | I | I | I | I | I |
2 | I | I | I | I | I | I | I | I | I |
3 | I | I | I | I | I | I | I | Wt | I |
4 | I | I | I | I | I | I | I | I | I |
5 | I | I | I | I | I | I | I | I | I |
Image test results of ice road surface under good illumination condition (from the experimental system). (a) Image of ice road surface state. (b) Image blocks.
Table
The ice road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 45 | 0 | 0 | 0 | 0 |
Wet | Wt | 3 | 1 | 6.67% | 2.22% | |
Water | Wr | 0 | 0 | 0 | 0 | |
Snow | S | 0 | 0 | 0 | 0 | |
Ice | I | 42 | 44 | 93.33% | 97.78% |
From Table
The recognition results of the ice road surface state with snow (from the Internet image) are shown in Figure
Recognition results of grid.
Column | Row | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
1 | D | I | Wt | I | I | I |
2 | I | I | I | I | I | D |
3 | D | I | I | I | I | I |
4 | Wt | I | I | I | I | Wt |
5 | I | Wr | I | I | I | I |
6 | D | Wt | D | I | I | I |
7 | I | I | I | I | I | I |
8 | I | I | I | D | D | Wt |
Recognition results of PSO.
Column | Row | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
1 | I | I | I | I | S | I |
2 | I | I | I | S | I | I |
3 | I | I | I | S | S | I |
4 | I | I | I | I | I | I |
5 | I | I | I | I | I | Wt |
6 | I | Wt | Wt | I | I | I |
7 | I | I | I | I | Wt | Wt |
8 | I | I | I | I | I | Wt |
Image test results of ice road surface with snow (from the Internet image). (a) Image of ice road surface state. (b) Image blocks.
Table
The ice road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 45 | 42 | 45 | 93.33% | 100% |
Wet | Wt | 3 | 0 | 6.67% | 0 | |
Water | Wr | 0 | 0 | 0 | 0 | |
Snow | S | 0 | 0 | 0 | 0 | |
Ice | I | 0 | 0 | 0 | 0 |
From Table
The recognition results of the ice, wet, and water hybrid state (from the mobile image) are shown in Figure
Recognition results of grid.
Column | Row | |||||||
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
1 | Wt | D | Wr | Wr | Wr | Wr | D | D |
2 | Wt | Wt | D | Wr | Wr | Wr | D | Wt |
3 | Wt | Wt | D | Wr | Wr | Wr | D | Wt |
4 | Wt | Wt | Wr | Wt | Wr | Wt | D | Wt |
5 | S | S | Wt | Wt | Wt | Wt | D | Wt |
6 | D | Wt | Wr | Wt | D | D | D | D |
Recognition results of PSO.
Column | Row | |||||||
---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
1 | Wt | D | D | Wt | Wr | Wt | Wr | Wr |
2 | D | D | D | Wt | Wr | Wt | Wr | Wt |
3 | D | D | Wr | Wr | Wr | Wt | Wt | Wr |
4 | D | D | Wr | Wr | Wr | Wt | D | Wt |
5 | D | Wt | Wr | Wr | Wr | Wr | Wt | Wt |
6 | D | D | Wr | Wr | Wr | Wr | Wr | Wt |
Image test results of hybrid road surface (from the mobile image). (a) Image of hybrid road surface state. (b) Image blocks.
Table
The hybrid road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 48 | 14 | 13 | 29.17% | 27.08% |
Wet | Wt | 19 | 14 | 39.58% | 29.17% | |
Water | Wr | 13 | 21 | 27.08% | 43.75% | |
Snow | S | 2 | 0 | 0.42% | 0 | |
Ice | I | 0 | 0 | 0 | 0 |
From Table
The recognition results of the ice, wet, and water hybrid state (from the surveillance system) are shown in Figure
Recognition results of grid.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | Wt | Wr | I | Wr | Wt | I | I | I | S |
2 | Wt | I | I | I | Wt | I | I | Wt | S |
3 | Wt | Wr | S | Wr | Wt | I | Wt | Wt | S |
4 | Wt | Wr | I | Wr | Wt | Wt | I | Wr | Wr |
5 | Wt | Wr | I | Wr | Wt | I | Wr | Wr | S |
Recognition results of PSO.
Column | Row | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
1 | I | I | I | I | I | I | Wt | Wt | I |
2 | I | Wt | Wt | I | I | I | I | I | I |
3 | S | S | S | I | I | I | I | I | I |
4 | S | I | S | I | I | I | Wt | I | I |
5 | S | I | I | I | I | I | S | I | Wt |
Image test results of hybrid road surface (from the surveillance system). (a) Image of hybrid road surface state. (b) Image blocks.
Table
The hybrid road surface image identification results statistics.
State | State symbol | Number of blocks | Grid search | PSO | Grid search/% | PSO/% |
---|---|---|---|---|---|---|
Dry | D | 45 | 0 | 0 | 0 | 0 |
Wet | Wt | 14 | 6 | 31.11% | 13.33% | |
Water | Wr | 12 | 0 | 26.67% | 0 | |
Snow | S | 5 | 7 | 11.11% | 15.56% | |
Ice | I | 14 | 32 | 31.11% | 71.11% |
From Table
There are a large number of traffic accidents caused by bad weather condition or slippery road condition. Therefore, road states greatly affect the traffic safety and transport efficiency on highway. It is of great social significance to study the classification of wet and slippery road condition, which can provide reference and theoretical basis for traffic control and meteorological management and ensure traffic safety.
There are many limitations in using instrument to recognize road surface conditions, and image recognition is becoming the main technology for recognizing road surface state. However, recognition under hybrid road conditions and different lighting conditions are two problems that need to be solved.
Based on SVM algorithm and image segmentation processing technology, we propose a method of video image processing technology for road surface state recognition. First of all, according to the segmentation principle, the road surface samples are divided into blocks and the road surface state sample database is constructed. Then, 9-dimensional color eigenvectors and 4-dimensional texture eigenvectors are extracted to form a 13-dimensional eigenvectors database which can describe the road surface state. After that, the SVM classifier is trained by using grid searching optimization and PSO optimization to obtain the road surface state classification model. And then, the performances of two classification models are tested. Finally, a road surface state recognition program was developed to test the actual road surface state images in a variety of environments.
The test results show that (1) the establishment of a perfect sample database is the basis for accurate recognition of road surface state. The quality and purity of the sample database can be ensured by dealing with single state image blocks. (2) Each feature value of the five states has overlapping parts, while 13-dimensional eigenvectors can satisfy the need of state recognition accurately. (3) After the SVM parameter optimization, the performance of road state classification model is superior, in which the performance of the PSO algorithm is better than that of the grid searching optimization algorithm, and the accuracy of state recognition is improved. (4) Image segmentation method can be used to obtain the distribution of road surface state, which solves the problem of hybrid road surface state and road surface under different light conditions. The recognition accuracy of single state is above 90%, and the recognition accuracy of hybrid state is more than 85%.
The authors declare that there are no conflicts of interest regarding the publication of this article.
Jiandong Zhao and Hongqiang Wu presented the algorithms, analyzed the data, and cowrote the paper; Liangliang Chen installed the experimental system and performed the experiments.
This work is supported by the Fundamental Research Funds for the Central Universities (2016JBM053).