CN105469099A - Sparse-representation-classification-based pavement crack detection and identification method - Google Patents

Abstract

The invention provides a sparse-representation-classification-based pavement crack detection and identification method. A sparse representation-based classifier (SRC) is introduced and an effective sub-block high-order moment characteristic is selected, thereby avoiding pretreatment and post treatment of an image, simplifying detection steps, and improving the operation efficiency. The method comprises: carrying out extraction and normalization of training set (subblock) characteristic vectors; dividing a tested image into a plurality of sub blocks and extracting features of the sub blocks, and carrying out classification by using an SRC; and identifying a crack type according to a mapping code of a sub block classification result. Compared with the traditional crack detection and identification method, the provided method has the higher identification precision and higher execution efficiency.

Description

Pavement crack detection and identification method based on sparse representation classification

Technical Field

The invention relates to the field of computer vision and pattern recognition, mainly uses a machine learning method for detecting and recognizing pavement cracks, and particularly relates to a pavement crack detection and recognition method based on sparse representation classification.

Background

Cracks are the most common diseases of the road surface, and the timely and accurate discovery of the cracks of the road surface is of great importance to the maintenance management of the high-load road. Through artificial vision detection, a large amount of manpower and material resources are needed, and the detection result has human subjectivity. The rapid development of computers enables people to use computers to complete automatic detection of pavement diseases.

The traditional pavement crack detection is based on image processing and analysis, and then cross-field methods are proposed to depict and enhance crack characteristics in complex environments and introduce a new idea for crack detection. Such as a method combined with fuzzy set theory, a detection strategy based on artificial population, a detection algorithm using a target point minimum spanning tree, and an algorithm based on fractional order differentiation. With the rapid development of pattern recognition technology, the method of machine learning is also applied to the detection and recognition of pavement cracks.

Pattern recognition, also commonly referred to as pattern classification, is classified into two categories, supervised classification (superviredisc classification) and unsupervised classification (unsupervised classification), from the viewpoint of the nature of the processing problem and the method of solving the problem. The main difference between the two is whether the category of each experimental sample is known in advance; in general, supervised classification often requires the provision of a large number of samples of known classes.

In recent years, various pattern recognition methods are applied to detection and recognition of pavement cracks, and due to the fact that noise components of actually collected pavement images are complex, a plurality of methods need to be preprocessed to eliminate the influence of partial noise, so that not only are the steps complex and the execution efficiency low, but also the pattern recognition effect greatly depends on image preprocessing.

Disclosure of Invention

The invention aims to provide a pavement crack detection and identification method based on sparse representation classification, so as to overcome the defects of low detection precision and long time consumption of the traditional method.

The above object of the invention is achieved by the features of the independent claims, the dependent claims developing the features of the independent claims in alternative or advantageous ways.

In order to achieve the purpose, the invention provides a pavement crack detection and identification method based on sparse representation classification, which comprises the following steps of:

1) selecting a training set, extracting a feature vector set of the training set and normalizing;

2) labeling the sub-blocks of the training set, and dividing the sub-blocks into non-crack sub-blocks and crack sub-blocks;

3) dividing the test picture into subblocks with specified sizes, extracting the characteristics of each subblock and normalizing;

4) classifying each sub-block of the image by using SRC according to the characteristic vector of the test set to obtain sub-blocks

A type matrix of (2);

5) carrying out horizontal and longitudinal mapping coding on the sub-block type matrix and carrying out coding enhancement;

6) and identifying the crack type by using the code processed in the step 5).

It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be considered as part of the inventive subject matter of this disclosure unless such concepts are mutually inconsistent. In addition, all combinations of claimed subject matter are considered a part of the presently disclosed subject matter.

The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description taken in conjunction with the accompanying drawings. Additional aspects of the present invention, such as features and/or advantages of exemplary embodiments, will be apparent from the description which follows, or may be learned by practice of specific embodiments in accordance with the teachings of the present invention.

Drawings

The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. Embodiments of various aspects of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 is a sparse representation classification-based pavement crack detection and identification flow diagram of certain embodiments of the present invention.

FIG. 2 is a fracture identification flow diagram of certain embodiments of the invention.

Fig. 3 is a diagram of example mapping coding and coding enhancements for a sub-block classification matrix in accordance with some embodiments of the present invention. The middle is a sub-block type matrix, the thickened part in the middle is vertical coding, the character style is reduced to be horizontal coding, and the thickened part in the outermost periphery is new coding after coding enhancement.

Fig. 4 is a schematic diagram showing the crack recognition effect, "o" indicates a deviated longitudinal crack, "x" indicates a reticular crack, and "+" indicates a deviated transverse crack.

Fig. 5 is a schematic view of the fracture type.

Detailed Description

In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.

In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to include all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, as the disclosed concepts and embodiments are not limited to any one implementation. In addition, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.

According to the pavement crack detection and identification method based on sparse representation classification, a sparse representation classifier is introduced, and the high-order moment features of image subblocks are used as the basis of classification of the classifier, so that preprocessing of images is not needed, and execution efficiency and identification accuracy are greatly improved. The method mainly comprises three parts of extracting and normalizing the characteristic vector of a training set (sub-block), extracting the characteristic of a test image in a blocking mode, classifying the test image by using SRC, and identifying the crack type according to the mapping coding of the sub-block classification result.

Some exemplary embodiments of the invention are described below with reference to the accompanying drawings.

According to the embodiment of the invention, a rapid pedestrian detection method based on similarity estimation is used for overcoming the problem that the existing pedestrian detection method based on a sliding window is too slow in detection speed. As shown in connection with fig. 1, the implementation of the method generally comprises the following 6 steps:

1) selecting a training set, extracting a feature vector set of the training set and normalizing;

2) labeling the sub-blocks of the training set, and dividing the sub-blocks into non-crack sub-blocks and crack sub-blocks;

3) dividing the test picture into subblocks with specified sizes, extracting the characteristics of each subblock and normalizing;

4) classifying each sub-block of the image by using SRC according to the characteristic vector of the test set to obtain sub-blocks

A type matrix of (2);

5) carrying out horizontal and longitudinal mapping coding on the sub-block type matrix and carrying out coding enhancement;

6) and identifying the crack type by using the code processed in the step 5).

In the above method, the step 1) specifically comprises:

11) selecting a plurality of pictures from the collected pictures to be divided into subblocks of n multiplied by n with specified sizes, wherein the subblocks are divided into crack subblocks and non-crack subblocks;

when the sub-blocks are too large, local features (fine cracks) may be ignored, resulting in sub-blocks containing cracks not being detected and a reduced recall rate. When the sub-blocks are too small, except that the processed data volume is too large and the efficiency is reduced, some noises can be misjudged as cracks, so that the accuracy rate is reduced; to ensure accuracy and recall, the subblock size was set to 75 x 75.

12) Selecting M (M is more than or equal to 100) subblocks from the subblocks in the step 11) as a training set, extracting a feature vector (stdM3M4) of the subblock, wherein std is a standard deviation, and M3 and M4 are a third moment feature and a fourth moment feature of the subblock image;

the moment features are used as the features of the classifier, so that the classifier is quite robust to noise, preprocessing is not needed, and the operation efficiency is improved.

13) Normalizing the feature vector of each sub-block;

in the above method, the step 12) is specifically:

121) assuming that the sub-block is l (n × n), the standard deviation feature std of the sub-block is extracted as formula (1);

$std=\sqrt{\frac{{\mathrm{\Σ}}_{i=1}^n-{\mathrm{\Σ}}_{j=1}^n{\left(l\right(i,j)-means)}^2}{n^2-1}}---\left(1\right)$

wherein, $means=\frac{{\mathrm{\Σ}}_{i=1}^n{\mathrm{\Σ}}_{j=1}^{n}l(i,j)}{n^2};$

123) the moment feature extraction method of the sub-blocks is shown as formula (2), and when k is 3, the sub-blocks are third-order moment features, and when k is 4, the sub-blocks are fourth-order moment features.

$m_k=\frac{{\mathrm{\Σ}}_{i=1}^n{\mathrm{\Σ}}_{j=1}^n{\left(l\right(i,j)-means)}^k}{n^2}---\left(2\right)$

In the above method, the step 2) is specifically:

21) labeling the sub-blocks of the test set, and dividing the sub-blocks into non-crack sub-blocks and crack sub-blocks;

22) the labels of the sub-blocks of each test set are set to 0 and 1 according to the labeled grountruth, with 0 representing a non-fractured sub-block and 1 representing a fractured sub-block.

5. The pavement crack detection and identification method based on sparse representation classification as claimed in claim 4, wherein the step 3) specifically comprises the following steps:

31) dividing the test picture P (W × L) into subblocks of a specified size, the subblock size being the same as that in step 11), and obtaining K × N subblocks (M ═ W/N, N ═ L/N);

32) extracting a feature composition vector (stdM3M4) of each sub-block;

33) normalizing the feature vector;

in the above method, the step 4) is specifically:

41) for different types of pavements, extracting a feature vector set of a training set according to the step 1);

42) the classification of K sub-blocks of the test picture is done using SRC.

43) Obtaining a subblock type matrix p (M × N) according to the classification result in the step 42), which represents the classification condition of each subblock of the test picture, wherein p (i, j) (i is 0,1 … … M; j is 0,1 … … N) is 0 or 1,0 denotes a non-split sub-block, and 1 denotes a split sub-block.

As shown in fig. 2, in the above method, the step 5) specifically includes:

51) mapping and encoding the subblock type matrix p (M × N) in the step 43) in the horizontal direction and the vertical direction.

x_iIndicates the number of i-th column 1 in the subblock type matrix, y_jIndicating the number of 1's in the jth row of the subblock type matrix. As shown in the example of fig. 3;

52) with x_iFor example, a range X ═ X is selected_i-dx_i+d]And d is an enhancement distance parameter, and coding enhancement is carried out according to a formula (3) to obtain a new horizontal code.

x_i＝max(X)-min(X)(3)

As shown in fig. 3, d in the example is 1;

53) the vertical coding enhancement method is similar to horizontal coding.

As shown in fig. 4, in the above method, the step 1) specifically includes:

61) counting the number sum of sub-block labels of 1;

62) if sum is 0, the image type is a normal road surface, no crack exists, a result is output, and the method is ended;

63) if sum is not 0, finding the standard deviations stdx and stdy of the horizontal encoding and the vertical encoding, respectively, and performing step 64) and step 65);

64) calculating the included angle theta between the two values and the horizontal axis in the rectangular coordinate system through stdx and stdy, as shown in a formula (4);

$\mathrm{\θ}=\arcsin (stdy/\sqrt{{\mathrm{stdy}}^2+{\mathrm{stdx}}^2})---\left(4\right)$

65) and (5) identifying the type of the crack of the image according to a formula (5).

The preferential cracks indicate that there are many cracks in this direction, while the net cracks have no significant bias. As shown in FIG. 6, the first three figures are off-set transverse slits and the last figure is a web slit.

In summary, the invention provides a pavement crack detection and identification method based on sparse representation classification. Aiming at the problem that the defects of low detection precision and long time consumption exist generally when the machine learning method is applied to the detection and identification of the pavement cracks, a Sparse Representation Classifier (SRC) is introduced, and by selecting effective sub-block high-order moment characteristics, the preprocessing and post-processing of images are avoided, the detection steps are simplified, and the operation efficiency is improved. The method specifically comprises the following steps: extracting and normalizing the characteristic vector of the training set (sub-block), partitioning the test image to extract the characteristic, classifying by using SRC, and identifying the crack type according to the mapping coding of the sub-block classification result. Compared with the traditional crack inspection vehicle identification method, the method provided by the invention has higher identification precision and execution efficiency.

Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.

Claims (8)

1. A pavement crack detection and identification method based on sparse representation classification is characterized by comprising the following steps:

1) selecting a training set, extracting a feature vector set of the training set and normalizing;

2) labeling the sub-blocks of the training set, and dividing the sub-blocks into non-crack sub-blocks and crack sub-blocks;

3) dividing the test picture into subblocks with specified sizes, extracting the characteristics of each subblock and normalizing;

4) classifying each sub-block of the image by using SRC according to the characteristic vector of the test set to obtain a type matrix of the sub-block;

5) carrying out horizontal and longitudinal mapping coding on the sub-block type matrix and carrying out coding enhancement;

6) and identifying the crack type by using the code processed in the step 5).

2. The sparse representation classification-based pavement crack detection and identification method according to claim 1, wherein the step 1) specifically comprises the following steps:

11) selecting a plurality of pictures from the collected pictures to be divided into subblocks of n multiplied by n with specified sizes, wherein the subblocks are divided into crack subblocks and non-crack subblocks;

12) selecting M subblocks with the number being more than or equal to 100 from the subblocks in the step 11) as a training set, extracting a feature vector (stdM3M4) of the subblocks, wherein std is a standard deviation, and M3 and M4 are third-order moment features and fourth-order moment features of subblock images;

13) the feature vectors of each sub-block are normalized.

3. The pavement crack detection and identification method based on sparse representation classification as claimed in claim 2, wherein the step 12) comprises the following steps:

121) assuming that the subblock is l and the size of the subblock is n multiplied by n, the method for extracting the standard deviation feature std of the subblock is as shown in the formula (1);

$std=\sqrt{\frac{{\mathrm{\Σ}}_{i=1}^n{\mathrm{\Σ}}_{j=1}^n{\left(l\right(i,j)-means)}^2}{n^2-1}}---\left(1\right)$

wherein, $means=\frac{{\mathrm{\Σ}}_{i=1}^n{\mathrm{\Σ}}_{j=1}^{n}l\left(i,j\right)}{n^2};$

122) the moment feature extraction method of the sub-blocks is as in formula (2), wherein when k is 3, the moment feature is a third-order moment feature, and when k is 4, the moment feature is a fourth-order moment feature:

$m_k=\frac{{\mathrm{\Σ}}_{i=1}^n{\mathrm{\Σ}}_{j=1}^n{\left(l\right(i,j)-means)}^k}{n^2}.---\left(2\right)$

4. the pavement crack detection and identification method based on sparse representation classification as claimed in claim 3, wherein the step 2) specifically comprises the following steps:

21) labeling the sub-blocks of the test set, and dividing the sub-blocks into non-crack sub-blocks and crack sub-blocks;

22) the labels of the sub-blocks of each test set are set to 0 and 1 according to the labeled grountruth, with 0 representing a non-fractured sub-block and 1 representing a fractured sub-block.

5. The pavement crack detection and identification method based on sparse representation classification as claimed in claim 4, wherein the step 3) specifically comprises the following steps:

31) dividing the test picture P with the size of W multiplied by L into subblocks with the specified size, wherein the subblock size is the same as that in the step 11), and obtaining K multiplied by N subblocks, M multiplied by W/N, and N multiplied by L/N;

32) extracting a feature composition vector (stdM3M4) of each sub-block;

33) and normalizing the feature vector.

6. The pavement crack detection and identification method based on sparse representation classification as claimed in claim 5, wherein the step 4) specifically comprises the following steps:

41) for different types of pavements, extracting a feature vector set of a training set according to the step 1);

42) finishing classification of K sub-blocks of the test picture by using the SRC;

43) obtaining a sub-block type matrix p (M × N) according to the classification result in step 42), which represents the classification condition of each sub-block of the test picture, where p (i, j) is 0 or 1,0 represents a non-crack sub-block, 1 represents a crack sub-block, and i is 0,1, 2.

7. The pavement crack detection and identification method based on sparse representation classification as claimed in claim 6, wherein the step 5) specifically comprises the following steps:

51) mapping and coding the subblock type matrix p (M × N) in the step 43) in the horizontal direction and the vertical direction to obtain horizontal codes (x)₁x₂……x_N) And vertical coding (y)₁y₂……y_M)；

x_iIndicates the number of i-th column 1 in the subblock type matrix, y_jIndicating the number of 1's in the jth row of the subblock type matrix.

52) For x_iSelecting a range of X ═ X_i-dx_i+d]D is an enhancement distance parameter, and coding enhancement is carried out according to a formula (3) to obtain a new horizontal code:

x_i＝max(X)-min(X)(3)

53) vertical coding enhancement is performed in the same manner as horizontal coding described above.

8. The pavement crack detection and identification method based on sparse representation classification as claimed in claim 7, wherein the step 6) specifically comprises the following steps:

61) counting the number sum of sub-block labels of 1;

62) if sum is 0, the image type is a normal road surface, no crack exists, a result is output, and the method is ended;

63) if sum is not 0, finding the standard deviations stdx and stdy of the horizontal encoding and the vertical encoding, respectively, and performing step 64) and step 65);

64) and (3) calculating an included angle theta between the two values and a horizontal axis in the rectangular coordinate system through stdx and stdy, as shown in a formula (4):

$\mathrm{\θ}=\arcsin (stdy/\sqrt{{\mathrm{stdy}}^2+{\mathrm{stdx}}^2})---\left(4\right)$

65) and (3) identifying the type of the image according to a formula (5):