CN111832571B - Automatic detection method for truck brake beam strut fault - Google Patents
Automatic detection method for truck brake beam strut fault Download PDFInfo
- Publication number
- CN111832571B CN111832571B CN202010656792.3A CN202010656792A CN111832571B CN 111832571 B CN111832571 B CN 111832571B CN 202010656792 A CN202010656792 A CN 202010656792A CN 111832571 B CN111832571 B CN 111832571B
- Authority
- CN
- China
- Prior art keywords
- image
- brake beam
- beam strut
- fault
- feature points
- 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.)
- Active
Links
Images
Classifications
-
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T17/00—Component parts, details, or accessories of power brake systems not covered by groups B60T8/00, B60T13/00 or B60T15/00, or presenting other characteristic features
- B60T17/18—Safety devices; Monitoring
- B60T17/22—Devices for monitoring or checking brake systems; Signal devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Abstract
A method for automatically detecting the fault of a truck brake beam strut belongs to the technical field of truck brake beam strut fault detection. The invention solves the problems of missing detection and error detection of the brake beam strut faults easily caused by adopting a manual detection method. According to the invention, high-definition imaging equipment is respectively built around the rail of the truck, and the truck acquires a clear image after passing through the equipment. The brake beam strut members are coarsely positioned using wheel base information and a priori knowledge of the brake beam strut position. And further processing the image of the coarse positioning component to judge whether the image is broken. And if the image is broken, mapping the recognition result into the original image according to the mapping relation, and alarming and uploading. And the staff performs corresponding processing according to the identification result to ensure the safe operation of the train. The invention can be applied to the fault detection of the brake beam strut of the truck.
Description
Technical Field
The invention belongs to the technical field of truck brake beam strut fault detection, and particularly relates to an automatic truck brake beam strut fault detection method.
Background
The brake beam strut is a main accessory of the brake beam, and if the brake beam strut is broken off and cannot be found and maintained in time, cracks become larger and larger, so that a lower pull rod falls off a steel rail to cause dangerous accidents, and serious accidents such as vehicle overturn, line disconnection and the like can be caused in serious cases, and the consequences are not considered. The brake beam strut breakage fault is a fault which endangers driving safety, and in the brake beam strut fault detection, the fault detection is usually carried out in a mode of manually checking images in the existing method. In the detection process, the detection result is influenced by subjective factors of vehicle detection personnel, so that the problems of missed detection, wrong detection and the like of faults are easily caused, and the driving safety is influenced.
Disclosure of Invention
The invention aims to solve the problems of missed detection and false detection of the brake beam strut faults easily caused by adopting a manual detection method, and provides an automatic detection method for the brake beam strut faults of a freight car.
The technical scheme adopted by the invention for solving the technical problems is as follows: a truck brake beam strut fault automatic detection method specifically comprises the following steps:
step one, acquiring an original image containing a brake beam strut component;
step two, respectively carrying out gradient transformation, Laplace transformation and wavelet transformation on the original image to obtain an original image after the gradient transformation, an original image after the Laplace transformation and an original image after the wavelet transformation;
respectively extracting ORB characteristic points and characteristic vectors of an original image, an original image after gradient transformation, an original image after Laplace transformation and an original image after wavelet transformation, and taking all the extracted ORB characteristic points as a first group of ORB characteristic points;
thirdly, carrying out image fusion on the brake beam strut component in the image database to obtain a template image of the brake beam strut component; respectively carrying out gradient transformation, Laplace transformation and wavelet transformation on the template image to obtain a template image after the gradient transformation, a template image after the Laplace transformation and a template image after the wavelet transformation;
respectively extracting ORB characteristic points and characteristic vectors of the template image, the template image after gradient transformation, the template image after Laplace transformation and the template image after wavelet transformation, and taking all the extracted ORB characteristic points as a second group of ORB characteristic points;
step four, performing initial feature point matching on the first group of ORB feature points and the second group of ORB feature points according to the feature vectors extracted in the step two and the step three;
fifthly, removing the initially successfully matched feature point pairs meeting preset removing conditions to obtain residual feature point pairs;
step six, screening and purifying the remaining characteristic point pairs to obtain screened and purified characteristic point pairs;
seventhly, obtaining the fine position of the brake beam strut component in the original image according to the screened and purified characteristic point pairs;
step eight, carrying out similarity analysis on the finely positioned brake beam strut component image and the template image, if the similarity condition is not met, directly carrying out fault alarm on the original image, and otherwise, continuously executing the step nine;
step nine, carrying out histogram adjustment on the finely positioned brake beam strut component image to obtain a brake beam strut component image after the histogram adjustment;
step ten, processing the image obtained in the step nine to obtain a suspected fault area; then carrying out fault judgment on the suspected fault area;
step eleven, if the fault judgment result in the step eleven is that a fault occurs, mapping the fault position to the image acquired in the step one, performing fault alarm and performing fault display on a display interface; otherwise, the fault judgment result in the step ten is that no fault occurs, and the identification of the next image is started.
The invention has the beneficial effects that: the invention provides an automatic detection method for a truck brake beam strut fault, which combines the edge corner and corner distinguishing characteristics of a component in a fine positioning process, fuses original images, gradient images, Laplace transform images and multi-scale edge image feature points after wavelet transform, and avoids the conditions of few extracted feature points and final positioning failure caused by more image noise while extracting key features of the component. In fault recognition, the brake beam strut component is finely positioned, so that misinformation of a non-component area is avoided, the recognition accuracy is high, and the condition of false detection is avoided.
The automatic image identification mode is used for replacing a manual detection mode, so that the fault detection efficiency can be improved, and the problem that the manual detection method is easy to miss detection is solved.
Drawings
FIG. 1 is a flow chart of a method for automatically detecting a truck brake beam strut failure in accordance with the present invention;
FIG. 2 is a flow chart of the present invention for fine positioning of the brake beam strut position;
FIG. 3 is a flow chart of image brightness normalization according to the present invention.
Detailed Description
The first embodiment is as follows: the present embodiment will be described with reference to fig. 1 to 3. The method for automatically detecting the fault of the truck brake beam strut specifically comprises the following steps:
step one, acquiring an original image containing a brake beam strut component;
step two, respectively carrying out gradient transformation, Laplace transformation and wavelet transformation on the original image to obtain an original image after the gradient transformation, an original image after the Laplace transformation and an original image after the wavelet transformation;
respectively extracting ORB characteristic points and characteristic vectors of an original image, an original image after gradient transformation, an original image after Laplace transformation and an original image after wavelet transformation, and taking all the extracted ORB characteristic points as a first group of ORB characteristic points;
other parts of the non-brake beam support column part are easy to interfere with the identification of the break fault of the brake beam support column, and the method can be used for further fault identification after finely positioning the support column. Feature extraction is carried out on the original image, the gradient image, the Laplace transform image and the multi-scale edge image after wavelet transform, the point logarithm of the final successful matching can be improved, and the component positioning precision is high;
thirdly, carrying out image fusion on the brake beam strut component in the image database to obtain a template image of the brake beam strut component; respectively carrying out gradient transformation, Laplace transformation and wavelet transformation on the template image to obtain a template image after the gradient transformation, a template image after the Laplace transformation and a template image after the wavelet transformation;
respectively extracting ORB characteristic points and characteristic vectors of the template image, the template image after gradient transformation, the template image after Laplace transformation and the template image after wavelet transformation, and taking all the extracted ORB characteristic points as a second group of ORB characteristic points;
step four, performing initial feature point matching on the first group of ORB feature points and the second group of ORB feature points according to the feature vectors extracted in the step two and the step three;
fifthly, removing the initially successfully matched feature point pairs meeting preset removing conditions to obtain residual feature point pairs;
step six, screening and purifying the remaining characteristic point pairs to obtain screened and purified characteristic point pairs;
seventhly, obtaining the fine position of the brake beam strut component in the original image according to the screened and purified characteristic point pairs;
step eight, carrying out similarity analysis on the finely positioned brake beam strut component image and the template image, if the similarity condition is not met, directly carrying out fault alarm on the original image, and otherwise, continuously executing the step nine;
the condition that the similarity degree is not satisfied means that: the maximum similarity of the normalized correlation coefficient match is less than T1 and the maximum similarity of the normalized correlation coefficient match is less than T2;
step nine, carrying out histogram adjustment on the finely positioned brake beam strut component image to obtain a brake beam strut component image after the histogram adjustment;
and step nine, standardizing the brightness of the image. The finely positioned strut images are subjected to histogram adjustment according to the template histogram, the image histogram has the same shape as the template histogram, the adjusted brake beam strut images can keep the image brightness and contrast consistent, and the problem that the missed reports are easily caused due to the fact that the image brightness and contrast are different at different stations is solved.
Step ten, processing the image obtained in the step nine to obtain a suspected fault area; then carrying out fault judgment on the suspected fault area;
step eleven, if the fault judgment result in the step eleven is that a fault occurs, mapping the fault position to the image acquired in the step one, performing fault alarm and performing fault display on a display interface; otherwise, the fault judgment result in the step ten is that no fault occurs, and the identification of the next image is started.
In the fine positioning process, the characteristic that edge edges and corners of the component are clear is combined, the original image, the gradient image, the Laplace transform image and the multi-scale edge image feature points after wavelet transform are fused, and the situations that few feature points are extracted and final positioning fails due to more image noise are avoided while the key features of the component are extracted. In fault recognition, the brake beam strut component is finely positioned, so that misinformation of a non-component area is avoided, and the recognition accuracy is high.
The method combines the part characteristics of the brake beam strut, utilizes the straight line between the maximum connected region obtained by carrying out primary image processing such as binarization, morphological processing, extraction of the outline of the connected region, Hough transform and the like on the brake beam strut and the upper and lower layers of struts as a reference, can effectively eliminate mismatching points, purify the matching characteristic pair and improve the positioning accuracy and precision.
The invention applies the image brightness standardization algorithm to the automatic brake beam support fault identification, the requirements of the whole fault identification algorithm on brightness and contrast are reduced, and false alarm of fault identification is reduced.
The second embodiment is as follows: the first difference between the present embodiment and the specific embodiment is: the specific process of the step one is as follows:
step one, acquiring a truck image;
and step two, roughly positioning a subgraph of the position of the brake beam strut member from the truck image obtained in the step one by using wheel base information and priori knowledge of the position of the brake beam strut member, and taking the roughly positioned subgraph as an original image containing the brake beam strut member.
The third concrete implementation mode: the second embodiment is different from the first embodiment in that: the specific method for acquiring the truck image in the steps comprises the following steps: and (4) building high-definition imaging equipment around the rail of the truck, and acquiring the image of the truck when the truck passes through the high-definition imaging equipment.
The fourth concrete implementation mode: the second embodiment is different from the first embodiment in that: the wheel base information is generated through calculation of the near-end magnetic steel and the far-end magnetic steel.
According to the invention, the wheel base information generated by hardware is directly utilized, the subsequent fine positioning step can make up for the problem of large error of the wheel base information, and the overall fault identification speed of the system is high and the efficiency is high.
The fifth concrete implementation mode: the first difference between the present embodiment and the specific embodiment is: in the fourth step, the primary matching of the feature points is performed on the ORB feature points extracted in the second step and the third step, and the method adopted for the primary matching of the feature points is a KNN method (K nearest neighbor method).
The invention adopts the KNN method to carry out preliminary matching, and can still quickly finish the preliminary matching even under the condition of a plurality of fusion feature points.
Under the influence of different stations, hardware parameters, time, camera angles and imaging environments, the brake beam strut components have differences in image brightness, local deformation, angles, dimensions and the like. The method comprises the steps of firstly carrying out image fusion on fine brake beam strut components in an image database to obtain a template image of the fine strut components. The fused image has more and more valuable information. The brake beam strut component has clear edges and corners, and still has obvious boundaries after image transformation, so that the mismatching of the feature points can be avoided subsequently based on the feature, and the invention uses knn algorithm to carry out the initial matching of the feature points on all the feature points.
The matching feature pairs obtained by the initial matching have a plurality of wrong matches, which easily causes errors in positioning results. The component image change is large in the vehicle traveling direction, and the component position change is small in the vertical vehicle traveling direction. The proportion of the correct matching points in the current coarse positioning image is based on the proportion of the feature points in the template image in the template. According to the priori knowledge, the method can remove matched error points, such as error matched points with coordinates which are in or out of the image in a serious proportion and crossed, for the first time, so that the positioning accuracy can be enhanced while the efficiency is considered. In view of the influence of various factors such as image water stain, white paint, noise of a camera sensor and the like, mismatching still exists in the matching feature points. The invention screens and purifies again according to the relationship between the characteristic points and the maximum communication area and the straight line (for example, the matched points cannot be removed in the communication area or outside the communication area at the same time, and the matched points cannot be removed on the straight line or one side of the straight line at the same time), eliminates wrong matching points, improves the matching accuracy and achieves robust matching. Finally, the fine position of the part in the image to be identified can be calculated by the matching matrix.
Due to the influence of severe weather factors such as rain and snow, in rare cases, the brake beam strut can be positioned wrongly. Therefore, similarity analysis needs to be performed on the accurately positioned component and the template image, and if the similarity is lower than a certain threshold, a fault alarm is directly performed.
The sixth specific implementation mode: the first difference between the present embodiment and the specific embodiment is: the concrete process of the step five is as follows:
for pairs of feature points (a1, B1) for which the initial matching was successful, where a1 is a feature point in the first group of ORB feature points and B1 is a feature point in the second group of ORB feature points, if a1 is on the left side of the corresponding image and B1 is on the right side of the corresponding image, the pairs of feature points (a1, B1) are rejected, and if a1 is on the right side of the corresponding image and B1 is on the left side of the corresponding image, the pairs of feature points (a1, B1) are also rejected;
for pairs of feature points (a, B) and (a, B) that are successfully initially matched, where a and a are both feature points in a first set of ORB feature points, B and B are both feature points in a second set of ORB feature points, pairs of feature points (a, B) and (a, B) are rejected if a is on the left side of a and B is on the right side of B in the horizontal direction of the corresponding image, pairs of feature points (a, B) and (a, B) are also rejected if a is on the right side of a and B is on the left side of B in the horizontal direction of the corresponding image, pairs of feature points (a, B) and (a, B) are rejected if a is above a and B is below B in the vertical direction of the corresponding image, pairs of feature points (a, B) and (a, B) are also rejected if a is below a and B is above B in the vertical direction of the corresponding image.
The seventh embodiment: the first difference between the present embodiment and the specific embodiment is: in the sixth step, the remaining characteristic point pairs are screened and purified, and the specific process is as follows:
for any remaining pairs of feature points (a4, B4), where a4 is a feature point in the first group of ORB feature points and B4 is a feature point in the second group of ORB feature points, if a4 is within the maximum connected region of the corresponding image and B4 is not within the maximum connected region of the corresponding image, then rejecting the pair of feature points (a4, B4), and if a4 is not within the maximum connected region of the corresponding image and B4 is within the maximum connected region of the corresponding image, rejecting the pair of feature points (a4, B4);
if the characteristic point A4 is not on the same side of the same straight line as the characteristic point B4, rejecting a characteristic point pair (A4, B4);
and screening and purifying each remaining characteristic point pair to obtain screened and purified characteristic point pairs.
The feature point a4 being on the same side of the line as the feature point B4 means: a4 and B4 are distributed on two sides of the same straight line. When a4 and B4 are both on the same line, one of a4/B4 is on the same line, or a4 and B4 are on the same side of the same line, then feature point pairs (a4, B4) are not rejected, and feature point pairs (a4, B4) are rejected only when a4 and B4 are distributed on different sides of the same line.
The straight lines in the present embodiment are two straight lines at the upper edge and the lower edge of the brake beam stay member, and the feature point pairs are screened by the two straight lines, respectively.
The specific implementation mode is eight: the first difference between the present embodiment and the specific embodiment is: the concrete process of the seventh step is as follows:
calculating a matching matrix according to the screened and purified characteristic point pairs, and obtaining an upper left corner vertex coordinate, an upper right corner vertex coordinate, a lower left corner vertex coordinate and a lower right corner vertex coordinate of the brake beam strut component in the template image according to the matching matrix;
and respectively obtaining the coordinates of corresponding points of the top left corner vertex, the top right corner vertex, the bottom left corner vertex and the bottom right corner vertex of the brake beam strut component in the template image in the original image, and obtaining the fine position of the brake beam strut component in the original image.
The specific implementation method nine: the first difference between the present embodiment and the specific embodiment is: the concrete process of the ninth step is as follows:
and calculating a histogram template of the brake beam strut by combining fault databases of different stations, and performing histogram adjustment on the finely positioned brake beam strut component image according to the histogram template to obtain the histogram-adjusted brake beam strut component image.
The detailed implementation mode is ten: the first difference between the present embodiment and the specific embodiment is: processing the image obtained in the ninth step to obtain a suspected fault area, wherein the specific process is as follows:
carrying out image enhancement on the image obtained in the ninth step, and carrying out self-adaptive binarization segmentation to obtain a self-adaptive binarization segmented image;
performing morphological image transformation and connected region analysis on the segmented image to obtain a suspected fault region;
in the embodiment, the image self-adaptive binarization is adopted for segmentation, so that the algorithm robustness is improved.
The method for judging the fault of the suspected fault area specifically comprises the following steps:
and if the shape of the suspected fault area is a slender crack, the length of the suspected fault area penetrates through the support and the width of the suspected fault area is between 8 and 18 pixels, determining that the fault occurs, and otherwise, determining that the fault does not occur.
Since there is a crack penetrating the strut in the image after the brake beam strut is broken, and both sides of the crack are separated by a predetermined distance by the force, the failure point is a long and thin crack, and therefore this embodiment is used as a limiting condition for failure determination.
The above-described calculation examples of the present invention are merely to explain the calculation model and the calculation flow of the present invention in detail, and are not intended to limit the embodiments of the present invention. It will be apparent to those skilled in the art that other variations and modifications of the present invention can be made based on the above description, and it is not intended to be exhaustive or to limit the invention to the precise form disclosed, and all such modifications and variations are possible and contemplated as falling within the scope of the invention.
Claims (10)
1. A truck brake beam strut fault automatic detection method is characterized by comprising the following steps:
step one, acquiring an original image containing a brake beam strut component;
step two, respectively carrying out gradient transformation, Laplace transformation and wavelet transformation on the original image to obtain an original image after the gradient transformation, an original image after the Laplace transformation and an original image after the wavelet transformation;
respectively extracting ORB characteristic points and characteristic vectors of an original image, an original image after gradient transformation, an original image after Laplace transformation and an original image after wavelet transformation, and taking all the extracted ORB characteristic points as a first group of ORB characteristic points;
thirdly, carrying out image fusion on the brake beam strut component in the image database to obtain a template image of the brake beam strut component; respectively carrying out gradient transformation, Laplace transformation and wavelet transformation on the template image to obtain a template image after the gradient transformation, a template image after the Laplace transformation and a template image after the wavelet transformation;
respectively extracting ORB characteristic points and characteristic vectors of the template image, the template image after gradient transformation, the template image after Laplace transformation and the template image after wavelet transformation, and taking all the extracted ORB characteristic points as a second group of ORB characteristic points;
step four, performing initial feature point matching on the first group of ORB feature points and the second group of ORB feature points according to the feature vectors extracted in the step two and the step three;
fifthly, removing the initially successfully matched feature point pairs meeting preset removing conditions to obtain residual feature point pairs;
step six, screening and purifying the remaining characteristic point pairs to obtain screened and purified characteristic point pairs;
seventhly, obtaining the fine position of the brake beam strut component in the original image according to the screened and purified characteristic point pairs;
step eight, carrying out similarity analysis on the finely positioned brake beam strut component image and the template image, if the similarity condition is not met, directly carrying out fault alarm on the original image, and otherwise, continuously executing the step nine;
step nine, carrying out histogram adjustment on the finely positioned brake beam strut component image to obtain a brake beam strut component image after the histogram adjustment;
step ten, processing the image obtained in the step nine to obtain a suspected fault area; then carrying out fault judgment on the suspected fault area;
step eleven, if the fault judgment result in the step eleven is that a fault occurs, mapping the fault position to the image acquired in the step one, performing fault alarm and performing fault display on a display interface; otherwise, the fault judgment result in the step ten is that no fault occurs, and the identification of the next image is started.
2. The method for automatically detecting the failure of the brake beam strut of the wagon as claimed in claim 1, wherein the specific process of the first step is as follows:
step one, acquiring a truck image;
and step two, roughly positioning subgraphs of the positions of the brake beam strut members from the truck images acquired in the step one by utilizing the wheel base information and the priori knowledge of the positions of the brake beam strut members, and taking the roughly positioned subgraphs as original images containing the brake beam strut members.
3. The method for automatically detecting the fault of the truck brake beam strut as claimed in claim 2, wherein the specific method for acquiring the truck image in the step one is as follows: and (4) building high-definition imaging equipment around the rail of the truck, and acquiring the image of the truck when the truck passes through the high-definition imaging equipment.
4. The method as claimed in claim 2, wherein the wheel base information is calculated and generated from the proximal magnetic steel and the distal magnetic steel.
5. The method for automatically detecting the fault of the brake beam strut of the freight car as claimed in claim 1, wherein in the fourth step, the primary matching of the characteristic points is performed on the ORB characteristic points extracted in the second step and the third step, and the method adopted for the primary matching of the characteristic points is a KNN method.
6. The method for automatically detecting the failure of the brake beam strut of the wagon as claimed in claim 1, wherein the specific process of the fifth step is as follows:
for pairs of feature points (a1, B1) for which the initial matching was successful, where a1 is a feature point in the first group of ORB feature points and B1 is a feature point in the second group of ORB feature points, if a1 is on the left side of the corresponding image and B1 is on the right side of the corresponding image, the pairs of feature points (a1, B1) are rejected, and if a1 is on the right side of the corresponding image and B1 is on the left side of the corresponding image, the pairs of feature points (a1, B1) are also rejected;
for pairs of feature points (a, B) and (a, B) that are successfully initially matched, where a and a are both feature points in a first set of ORB feature points, B and B are both feature points in a second set of ORB feature points, pairs of feature points (a, B) and (a, B) are rejected if a is on the left side of a and B is on the right side of B in the horizontal direction of the corresponding image, pairs of feature points (a, B) and (a, B) are also rejected if a is on the right side of a and B is on the left side of B in the horizontal direction of the corresponding image, pairs of feature points (a, B) and (a, B) are rejected if a is above a and B is below B in the vertical direction of the corresponding image, pairs of feature points (a, B) and (a, B) are also rejected if a is below a and B is above B in the vertical direction of the corresponding image.
7. The method for automatically detecting the failure of the truck brake beam strut as claimed in claim 1, wherein in the sixth step, the remaining characteristic point pairs are screened and purified, and the specific process is as follows:
for any remaining pairs of feature points (a4, B4), where a4 is a feature point in the first group of ORB feature points and B4 is a feature point in the second group of ORB feature points, if a4 is within the maximum connected region of the corresponding image and B4 is not within the maximum connected region of the corresponding image, then rejecting the pair of feature points (a4, B4), and if a4 is not within the maximum connected region of the corresponding image and B4 is within the maximum connected region of the corresponding image, rejecting the pair of feature points (a4, B4);
if the characteristic point A4 is not on the same side of the same straight line as the characteristic point B4, rejecting a characteristic point pair (A4, B4);
and screening and purifying each remaining characteristic point pair to obtain screened and purified characteristic point pairs.
8. The automatic detection method for the fault of the truck brake beam strut as claimed in claim 1, wherein the specific process of the seventh step is as follows:
calculating a matching matrix according to the screened and purified characteristic point pairs, and obtaining an upper left corner vertex coordinate, an upper right corner vertex coordinate, a lower left corner vertex coordinate and a lower right corner vertex coordinate of the brake beam strut component in the template image according to the matching matrix;
and respectively obtaining the coordinates of corresponding points of the top left corner vertex, the top right corner vertex, the bottom left corner vertex and the bottom right corner vertex of the brake beam strut component in the template image in the original image, and obtaining the fine position of the brake beam strut component in the original image.
9. The method for automatically detecting the fault of the truck brake beam strut as claimed in claim 1, wherein the specific process of the ninth step is as follows:
and calculating a histogram template of the brake beam strut by combining fault databases of different stations, and performing histogram adjustment on the finely positioned brake beam strut component image according to the histogram template to obtain the histogram-adjusted brake beam strut component image.
10. The method for automatically detecting the failure of the truck brake beam strut as claimed in claim 1, wherein the image obtained in the ninth step is processed to obtain a suspected failure area, and the specific process is as follows:
carrying out image enhancement on the image obtained in the ninth step, and carrying out self-adaptive binarization segmentation to obtain a self-adaptive binarization segmented image;
performing morphological image transformation and connected region analysis on the segmented image to obtain a suspected fault region;
the method for judging the fault of the suspected fault area specifically comprises the following steps:
if the shape of the suspected defect area is a crack penetrating the pillar and the width of the suspected defect area is between 8 and 18 pixels, it is determined that a defect occurs, otherwise it is determined that no defect occurs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010656792.3A CN111832571B (en) | 2020-07-09 | 2020-07-09 | Automatic detection method for truck brake beam strut fault |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010656792.3A CN111832571B (en) | 2020-07-09 | 2020-07-09 | Automatic detection method for truck brake beam strut fault |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111832571A CN111832571A (en) | 2020-10-27 |
| CN111832571B true CN111832571B (en) | 2021-03-05 |
Family
ID=72900711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010656792.3A Active CN111832571B (en) | 2020-07-09 | 2020-07-09 | Automatic detection method for truck brake beam strut fault |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111832571B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112508908B (en) * | 2020-12-02 | 2021-07-16 | 哈尔滨市科佳通用机电股份有限公司 | Method for detecting disconnection fault of sanding pipe joint of motor train unit based on image processing |
| CN112651343B (en) * | 2020-12-28 | 2021-08-24 | 哈尔滨市科佳通用机电股份有限公司 | Railway wagon brake beam breaking fault identification method based on image processing |
| CN115170890B (en) * | 2022-07-28 | 2023-03-14 | 哈尔滨市科佳通用机电股份有限公司 | Method for identifying breakage fault of connecting pull rod chain of railway wagon |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008107303A1 (en) * | 2007-03-08 | 2008-09-12 | Cellavision Ab | A method for determining an in-focus position and a vision inspection system |
| CN104573713A (en) * | 2014-12-31 | 2015-04-29 | 天津弘源慧能科技有限公司 | Mutual inductor infrared image recognition method based on image textual features |
| CN110956601A (en) * | 2019-12-04 | 2020-04-03 | 山东大学 | Infrared image fusion method and device based on multi-sensor mode coefficients and computer readable storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005086137A1 (en) * | 2004-03-02 | 2005-09-15 | Beijing E-World Technology Co., Ltd. | A coding/decoding method based templet matching and multi-distinguishability analysis |
| CN105513038B (en) * | 2014-10-20 | 2019-04-09 | 网易（杭州）网络有限公司 | Image matching method and mobile phone application test platform |
| CN115049563A (en) * | 2015-12-31 | 2022-09-13 | 上海联影医疗科技股份有限公司 | Image processing method and system |
| CN106023185B (en) * | 2016-05-16 | 2019-07-30 | 国网河南省电力公司电力科学研究院 | A kind of transmission facility method for diagnosing faults |
-
2020
- 2020-07-09 CN CN202010656792.3A patent/CN111832571B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008107303A1 (en) * | 2007-03-08 | 2008-09-12 | Cellavision Ab | A method for determining an in-focus position and a vision inspection system |
| CN104573713A (en) * | 2014-12-31 | 2015-04-29 | 天津弘源慧能科技有限公司 | Mutual inductor infrared image recognition method based on image textual features |
| CN110956601A (en) * | 2019-12-04 | 2020-04-03 | 山东大学 | Infrared image fusion method and device based on multi-sensor mode coefficients and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111832571A (en) | 2020-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111832571B (en) | Automatic detection method for truck brake beam strut fault | |
| Hongliang et al. | A hybrid license plate extraction method based on edge statistics and morphology | |
| Hsieh et al. | Morphology-based license plate detection from complex scenes | |
| CN104966049B (en) | Lorry detection method based on image | |
| CN106156780B (en) | The method of wrong report is excluded on track in foreign body intrusion identification | |
| KR101717613B1 (en) | The moving vehicle detection system using an object tracking algorithm based on edge information, and method thereof | |
| CN103196372B (en) | A kind of optical imagery detection method of electrification railway contact net supportive device | |
| CN106548147A (en) | A kind of quick noise robustness image foreign matter detection method and TEDS systems | |
| CN111080601A (en) | Method for identifying fault image of pull ring grinding shaft of derailment brake device of railway wagon | |
| CN114758322B (en) | Road quality detection system based on machine identification | |
| CN113516629A (en) | Intelligent detection system for TFDS passing operation | |
| CN112287888A (en) | Track turning identification method based on prediction weight | |
| CN113077437A (en) | Workpiece quality detection method and system | |
| CN115527170A (en) | Method and system for identifying closing fault of door stopper handle of automatic freight car derailing brake device | |
| KR101902291B1 (en) | Traffic lane correction system and method thereof | |
| Skodras et al. | Rear lights vehicle detection for collision avoidance | |
| CN111832562B (en) | Spring supporting plate fault identification method based on image processing | |
| Zong et al. | Traffic light detection based on multi-feature segmentation and online selecting scheme | |
| CN112800974A (en) | Subway rail obstacle detection system and method based on machine vision | |
| CN109086671B (en) | Night lane marking line video detection method suitable for unmanned driving | |
| CN106803064A (en) | A kind of traffic lights method for quickly identifying | |
| CN112132824B (en) | Automatic detection method for failure of freight car axle box spring | |
| KR102585146B1 (en) | Pothole detection device and method based on deep learning | |
| CN109978879B (en) | Box corner in-groove state detection method based on railway wagon loading video monitoring | |
| CN108182376B (en) | Vehicle window corner detection-based vehicle window positioning method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |