CN112393617A - Method, device, equipment and medium for detecting wheels of trolley of annular cooler - Google Patents
Method, device, equipment and medium for detecting wheels of trolley of annular cooler Download PDFInfo
- Publication number
- CN112393617A CN112393617A CN202010223896.5A CN202010223896A CN112393617A CN 112393617 A CN112393617 A CN 112393617A CN 202010223896 A CN202010223896 A CN 202010223896A CN 112393617 A CN112393617 A CN 112393617A
- Authority
- CN
- China
- Prior art keywords
- detected
- video image
- trolley
- circular cooler
- wheels
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D21/00—Arrangements of monitoring devices; Arrangements of safety devices
- F27D21/02—Observation or illuminating devices
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D21/00—Arrangements of monitoring devices; Arrangements of safety devices
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D15/00—Handling or treating discharged material; Supports or receiving chambers therefor
- F27D15/02—Cooling
- F27D15/0206—Cooling with means to convey the charge
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D21/00—Arrangements of monitoring devices; Arrangements of safety devices
- F27D21/02—Observation or illuminating devices
- F27D2021/026—Observation or illuminating devices using a video installation
Abstract
The application discloses a method, a device, equipment and a medium for detecting wheels of a trolley of a circular cooler, which comprise the following steps: acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; when any frame of video image to be detected is acquired, inputting the video image to be detected to a pre-acquired post-training classifier; acquiring the predicted probability value of the current video image to be detected output by the trained classifier, wherein the current video image to be detected contains the wheels of the trolley of the circular cooler; judging whether the predicted probability value is smaller than a preset probability threshold value, if so, judging that the current video image to be detected does not contain wheels of the trolley of the circular cooler, and determining the time information corresponding to the current video image to be detected as target time; if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold, determining that the circular cooler trolley has a wheel drop fault; the duration takes the target time as an initial time. Wheel faults can be detected in time.
Description
Technical Field
The application relates to the technical field of application of ring cooling machines, in particular to a method, a device, equipment and a medium for detecting wheels of a trolley of a ring cooling machine.
Background
The function of the sintering circular cooler is to cool hot sintering ores discharged from a sintering machine, and the main structure of the sintering circular cooler comprises: trolley, rotary frame, transmission device, track, bellows and blower, etc. The trolley of the circular cooler is connected to the revolving frame through a hinge, and after receiving hot ores unloaded from the ore feeding funnel, the trolley makes circular motion along the horizontal circular track along the revolving frame, so that the horizontal circular track of the ore unloading area is changed into a curved track which is bent downwards, and the unloading operation is completed along the curved track trolley. The ring cooling trolley rotates ceaselessly according to the sintering amount, the equipment volume is large, the wheel load is heavy, the running environment of the wheel is severe, and the wheel is easy to damage, loosen and even fall off. And the circular cooler trolley moves circularly and has a large turning radius, so that the trolley is easy to deviate in the running process, the trolley rim is extruded and rubbed with a track, and the damage of the wheels is aggravated. After the wheels of the trolley fall off, accidents such as the inclination and the side turning of the trolley are easily caused, and the safety of crisis personnel is ensured.
Disclosure of Invention
In view of this, an object of the present application is to provide a method, an apparatus, a device and a medium for detecting wheels of a circular cooler pallet, which can detect whether a wheel drop fault occurs in the circular cooler pallet in time, so as to avoid occurrence of a safety accident. The specific scheme is as follows:
in a first aspect, the application discloses a method for detecting wheels of a trolley of a circular cooler, which comprises the following steps,
acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler;
when any frame of the video image to be detected is acquired, inputting the video image to be detected to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain the wheels of the trolley of the circular cooler and corresponding label information;
acquiring the predicted probability value of the current video image to be detected output by the trained classifier, wherein the current video image to be detected contains the wheels of the trolley of the circular cooler;
judging whether the prediction probability value is smaller than a preset probability threshold value, if so, judging that the current video image to be detected does not contain wheels of the trolley of the circular cooler, and determining the time information corresponding to the current video image to be detected as target time;
if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold, determining that the circular cooler trolley has a wheel drop fault; wherein the duration takes the target time as an initial time.
Optionally, the inputting the video image to be detected into a pre-acquired classifier after training includes,
traversing the video image to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected.
Optionally, the inputting the video image to be detected into a pre-acquired classifier after training includes,
intercepting a target to-be-detected area from the to-be-detected video image; the area of the target area to be detected is larger than the wheel area of the wheel of the trolley of the circular cooler;
traversing the target region to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected.
Optionally, the method for detecting the wheel of the circular cooler trolley further includes:
and determining the preset time threshold value by using the wheel distance of the adjacent circular cooler trolleys.
Optionally, the determining the preset time threshold by using the wheel distance between adjacent circular cooler trolleys includes:
if the video image to be detected is input to the pre-acquired classifier after training, including,
traversing the video image to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected,
determining the preset time threshold value as tThreshold value 1∈(0，Δtmin]；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
Optionally, the determining the preset time threshold by using the wheel distance between adjacent circular cooler trolleys includes:
if the video image to be detected is input into a pre-acquired classifier after training, intercepting a target region to be detected from the video image to be detected; the target area to be detected is larger than the wheel area of the wheels of the trolley of the circular cooler, the target area to be detected is traversed by utilizing a sliding window, the video images to be detected in the sliding window are sequentially input to a pre-acquired classifier after training, so that the classifier after training finishes image detection aiming at the video images to be detected,
determining the preset time threshold value as tThreshold value 2∈(Δtmin,Δtmax)；
Wherein the content of the first and second substances,
s is a ringAnd v is the running average speed of the circular cooler trolley.
Optionally, the method for detecting the wheel of the circular cooler trolley further includes:
extracting the features of the target training sample by using a preset feature extraction algorithm;
and training a preset classifier by using the target training sample after feature extraction to obtain the trained classifier.
In a second aspect, the application discloses a wheel detection device for a trolley of a circular cooler, comprising,
a target data acquisition module; the method comprises the steps of acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler;
the video image detection module is used for inputting the video image to be detected to a pre-acquired post-training classifier when any frame of the video image to be detected is acquired; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain the wheels of the trolley of the circular cooler and corresponding label information;
the detection result acquisition module is used for acquiring the prediction probability value of the current video image to be detected output by the trained classifier, which contains the wheels of the trolley of the circular cooler;
the first condition judgment module is used for judging whether the prediction probability value is smaller than a preset probability threshold value or not, and if the prediction probability value is smaller than the preset probability threshold value, judging that the current video image to be detected does not contain wheels of the trolley of the circular cooler;
the target time determining module is used for determining the time information corresponding to the current video image to be detected as target time under the condition that the first condition judging module judges that the current video image to be detected does not contain the wheels of the trolley of the circular cooler;
the second condition judgment module is used for judging that the wheel drop fault occurs in the circular cooler trolley if the duration of the acquired prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold; wherein the duration takes the target time as an initial time.
In a third aspect, the application discloses a wheel detection device of a circular cooler trolley, which comprises a processor and a memory; wherein the content of the first and second substances,
the memory is used for storing a computer program;
the processor is used for executing the computer program to realize the wheel detection method of the circular cooler trolley.
In a fourth aspect, the present application discloses a computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned method for wheel detection of a loop cooler trolley.
Therefore, the method and the device for detecting the video images acquire the video images to be detected and the time information corresponding to each frame of the video images to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler, and when any frame of the video image to be detected is acquired, the video image to be detected is input to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain wheels of the circular cooler trolley and corresponding label information, then a prediction probability value of the wheels of the circular cooler trolley contained in a current video image to be detected output by the classifier after training is obtained, then whether the prediction probability value is smaller than a preset probability threshold value or not is judged, if the prediction probability value is smaller than the preset probability threshold value, the current video image to be detected is judged not to contain the wheels of the circular cooler trolley, time information corresponding to the current video image to be detected is determined as target time, and if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold value, is larger than the preset time threshold value, wheel dropping failure of the circular cooler trolley is judged; wherein the duration takes the target time as an initial time. Therefore, the video images acquired in real time aiming at the wheels of the circular cooler trolley are used for detection, and whether the wheel falling fault occurs in the circular cooler trolley can be detected timely, so that the safety accident is avoided.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method for detecting wheels of a trolley of a circular cooler according to the present disclosure;
FIG. 2 is a schematic structural view of a trolley of a ring cooling machine disclosed in the present application;
FIG. 3 is a schematic view of a camera mounting location disclosed herein;
FIG. 4 is a flowchart of a particular method for detecting wheels of a trolley of a ring cooler according to the present disclosure;
FIG. 5 is a flowchart of a particular method for detecting wheels of a trolley of a ring cooler according to the present disclosure;
FIG. 6 is a schematic diagram of a sliding window search according to the present disclosure;
FIG. 7 is a diagram illustrating classification results of a trained classifier disclosed in the present application;
FIG. 8 is a flowchart of a particular method for detecting a wheel of a ring cooler pallet disclosed herein;
fig. 9 is a schematic diagram illustrating a search of a target to be detected region disclosed in the present application;
FIG. 10 is a schematic structural view of a wheel detecting device of a trolley of a circular cooler according to the present disclosure;
FIG. 11 is a schematic structural view of a ring cooler trolley wheel detection system as disclosed in the present application;
FIG. 12 is a schematic structural view of a wheel inspection apparatus for a trolley of a ring cooling machine according to the present disclosure;
fig. 13 is a diagram of a server structure disclosed in the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The ring cooling machine trolley rotates ceaselessly according to the sintering amount, the equipment volume is large, the wheel load is heavy, the running environment of the wheel is severe, and the wheel is easy to damage, loosen and even fall off. And the circular cooler trolley moves circularly and has a large turning radius, so that the trolley is easy to deviate in the running process, the trolley rim is extruded and rubbed with a track, and the damage of the wheels is aggravated. After the wheels of the trolley fall off, accidents such as the inclination and the side turning of the trolley are easily caused, and the safety of crisis personnel is ensured. Therefore, the wheel detection scheme of the circular cooler trolley is provided, whether the wheel falling fault of the circular cooler trolley occurs or not can be detected timely, and therefore safety accidents are avoided.
Referring to fig. 1, an embodiment of the present application discloses a method for detecting a wheel of a trolley of a ring cooling machine, including:
step S11: acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the circular cooler trolley.
Referring to fig. 2, fig. 2 is a schematic structural diagram of a trolley of a ring cooling machine disclosed in the present application. Wherein, 1 is the friction plate, 2 is the wheel, 3 is the outside track, 4 is the revolving rack, 5 is the platform truck, 6 is the side rail. As can be seen from figure 2, when the trolley deviates, the wheel rim can be extruded with the track, which aggravates the damage and loosening of the wheel, and even the wheel falls off.
Referring to fig. 3, fig. 3 is a schematic view of a mounting position of a camera disclosed in the present application, where 1 is a trolley of a circular cooler, 2 is a track of the circular cooler, and 3 is the camera. Video image acquisition is carried out on the wheels of the circular cooler trolley through the camera, and the camera mounting position is a space which can be stopped for processing in time when the trolley breaks down before entering the bend. The image acquisition devices can be installed by 1 set or multiple sets, namely, multiple cameras can be installed, and when multiple sets are installed, each image acquisition device can be numbered, so that the wheel falling position can be roughly positioned conveniently, for example, the wheel falling position is judged by the image acquired by the No. 1 camera and is close to the No. 1 camera. Fig. 3 is a schematic view of the installation positions of cameras when only one camera is installed on each of the two sides of the inner and outer rails, the installation positions of the cameras are not limited to the positions, and the installation number of the single-side cameras is not limited to one.
Step S12: when any frame of the video image to be detected is acquired, inputting the video image to be detected to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; and the second training sample is an image which does not contain the wheel of the circular cooler trolley and corresponding label information.
In a specific implementation manner, in the embodiment of the present application, a preset feature extraction algorithm may be used to extract features of the target training sample, and then the target training sample after feature extraction is used to train a preset classifier, so as to obtain a trained classifier. Specifically, in this embodiment, an annular cooler wheel image training library may be built first, an annular cooler trolley wheel area image is captured from a video image acquired of the annular cooler trolley wheels to obtain a first training sample, an image not including the annular cooler trolley wheels is acquired as a second training sample to obtain a target training sample, then a preset feature extraction algorithm may be used to perform feature extraction on the target training sample, then the annular cooler wheel image training library is generated by using the target training sample after feature extraction, and finally a classifier is trained through the training sample in the annular cooler wheel image training library. The adopted classifier training method can comprise a decision tree classification method, a naive Bayes classification algorithm, a classification algorithm based on a support vector machine, a neural network method, a k-nearest neighbor method and the like.
It should be noted that because the circular cooler runs continuously all the day, the illumination changes, and the interference caused by the local shadow and illumination change in the image can be effectively reduced by preprocessing the target training sample. That is, the present embodiment may perform feature extraction on an image first, thereby increasing robustness to illumination variation. The preset feature extraction algorithm may be HOG (Histogram of Oriented Gradient), SIFT (Scale-invariant feature transform), SURF (Speeded Up Robust Features), GABOR feature extraction, or the like.
In addition, in this embodiment, when any frame of the video image to be detected is acquired, feature extraction may be performed on the video image to be detected by using a preset feature extraction algorithm, and then the video image to be detected after feature extraction is input to a pre-acquired post-training classifier.
Step S13: and acquiring the predicted probability value of the current video image to be detected output by the trained classifier, which contains the wheels of the circular cooler trolley.
Step S14: and judging whether the prediction probability value is smaller than a preset probability threshold value, if so, judging that the current video image to be detected does not contain the wheels of the trolley of the circular cooler, and determining the time information corresponding to the current video image to be detected as the target time.
Step S15: if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold, determining that the circular cooler trolley has a wheel drop fault; wherein the duration takes the target time as an initial time.
And, after judging that cold quick-witted platform truck of ring appears falling the wheel trouble, this embodiment can generate corresponding alarm information, and will alarm information sends to target terminal to relevant personnel in time know the actual fault condition, in time handle, can specifically pop out corresponding video monitoring at target terminal, so that operating personnel more audio-visually see whether the wheel drops, thereby reduce the artifical inspection number of times that increases when the wrong report is alert. It can be understood that the detection process is more visual because the trolley of the video stream is subjected to wheel drop detection, and when the wheel drop alarm occurs, an operator can directly know the actual fault condition without going to the site to check or calling a site monitoring video record in addition.
In this embodiment, the preset time threshold may be determined by using a wheel distance between adjacent circular cooler trolleys.
Therefore, the method and the device for detecting the video images acquire the video images to be detected and the time information corresponding to each frame of the video images to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler, and when any frame of the video image to be detected is acquired, the video image to be detected is input to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain wheels of the circular cooler trolley and corresponding label information, then a prediction probability value of the wheels of the circular cooler trolley contained in a current video image to be detected output by the classifier after training is obtained, then whether the prediction probability value is smaller than a preset probability threshold value or not is judged, if the prediction probability value is smaller than the preset probability threshold value, the current video image to be detected is judged not to contain the wheels of the circular cooler trolley, time information corresponding to the current video image to be detected is determined as target time, and if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold value, is larger than the preset time threshold value, wheel dropping failure of the circular cooler trolley is judged; wherein the duration takes the target time as an initial time. Therefore, the video images acquired in real time aiming at the wheels of the circular cooler trolley are used for detection, and whether the wheel falling fault occurs in the circular cooler trolley can be detected timely, so that the safety accident is avoided.
For example, referring to fig. 4, fig. 4 is a specific method for detecting a wheel of a loop cooling machine trolley disclosed in an embodiment of the present application, which may acquire historical video data collected for the wheel of the loop cooling machine trolley, and then manually mark and extract a wheel region and a non-wheel region to obtain a target training sample. And performing feature extraction on the target training sample, then constructing a training sample library, and performing classifier training to obtain a trained classifier. And detecting the video image by using the trained classifier. And a timer can be implemented for calculating the duration of no wheel in the video image to be detected.
Referring to fig. 5, the application discloses a specific method for detecting wheels of a trolley of a circular cooler, which comprises the following steps,
step S21: acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler;
step S22: when any frame of the video image to be detected is acquired, traversing the video image to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected.
The trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; and the second training sample is an image which does not contain the wheel of the circular cooler trolley and corresponding label information.
In a specific implementation mode, traversal retrieval is performed by using a sliding window, and each frame of video image to be detected is searched. The sliding window is arranged to move from left to right and from top to bottom for retrieval. The sliding window is defined as Rect (x, y, width, height), wherein x and y are coordinate positions of the upper left corner point of the sliding window in the video image to be detected, and width is the width of the sliding window and the height of the height sliding window. The sliding interval of the sliding window is step, and the moving calculation mode of the sliding window is as follows:
xi+step,if(xi-1+step+Rect.width)＜frame.width
yi+step,if(yi-1+step+Rect.height)＜frame.height
the frame is a current frame video image to be detected, which is acquired from the video stream. The selection of step influences the detection speed, the smaller the step, the larger the number of sliding windows, the higher the search detection precision, and the larger the step, the smaller the number of sliding windows, and the lower the search detection precision. i is the serial number of the sliding window, for example, the first sliding window, x, y is 0,0, and then the next top left corner point coordinate of the sliding window is x + step, y. First from left to right and then to the far right, the y-axis direction moves one step (i.e., moves down), and the x returns to the position of coordinate 0, sliding row by row in sequence.
Referring to fig. 6, fig. 6 is a schematic diagram of a sliding window retrieval according to the present disclosure.
In a specific implementation, the camera of the present embodiment has only one wheel within the coverage of the field of view.
Step S23: and acquiring the predicted probability value of the to-be-detected video image in the sliding window output by the trained classifier, wherein the to-be-detected video image comprises the wheels of the trolley of the circular cooler.
For example, referring to fig. 7, fig. 7 is a schematic diagram illustrating a classification result of a trained classifier disclosed in the present application. The video images to be detected in the sliding windows are input into the trained classifier for detection, the predicted probability value that the video images to be detected corresponding to each window contain the wheels of the circular cooler trolley can be obtained, and the larger the probability value is, the more possible the probability value is the wheel region.
Step S24: judging whether the prediction probability value corresponding to each window in the current to-be-detected image is smaller than a preset probability threshold value, if the prediction probability value corresponding to each window is smaller than the preset probability threshold value, judging that the current to-be-detected video image does not contain wheels of the annular cooler trolley, and determining the time information corresponding to the current to-be-detected video image as target time. That is, the time corresponding to the video image to be detected of the current frame is determined as the target time.
If the prediction probability value corresponding to any window in the current image to be detected is larger than or equal to the preset probability threshold, the current video image to be detected is judged to contain the wheels of the trolley of the circular cooler. And when the probability value obtained by recognition is still smaller than the set threshold value after the whole frame image is searched, judging that the frame image does not contain wheels.
Step S25: if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold, determining that the circular cooler trolley has a wheel drop fault; wherein the duration takes the target time as an initial time.
In a specific implementation manner, in this embodiment, the detection is performed by traversing the entire frame of the image to be detected through a sliding window, and it may be determined that the preset time threshold is tThreshold value 1∈(0，Δtmin]；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
It will be appreciated that the time interval between two adjacent wheels passing the fixed position of the track is deltatminThat is, occurThe time interval of the fixed position in the video image is Δ tminWhen one wheel leaves the camera view field, the time of the next adjacent wheel entering the camera is related to the camera view field range, and therefore the preset time threshold t can be determined according to the camera view field rangeThreshold value 1∈(0，Δtmin]。
Referring to fig. 8, an embodiment of the present application discloses a specific method for detecting a wheel of a trolley of a ring cooling machine, including:
step S31, acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the circular cooler trolley.
Step S32, when any frame of video image to be detected is obtained, intercepting a target area to be detected from the video image to be detected; the area of the target area to be detected is larger than the wheel area of the wheel of the trolley of the circular cooler; traversing the target region to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected;
the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; and the second training sample is an image which does not contain the wheel of the circular cooler trolley and corresponding label information.
It can be understood that a target to-be-detected area is set in the camera view field, and only the target to-be-detected area searches for the wheels, so that the search range can be reduced, and the detection speed is increased.
For example, referring to fig. 9, fig. 9 is a schematic diagram illustrating a retrieval of a target region to be detected according to an embodiment of the present application.
And S33, acquiring the predicted probability value that the target to-be-detected area in the sliding window output by the trained classifier contains the wheels of the circular cooler trolley.
And step S34, judging whether the prediction probability value corresponding to each window in the current target detection area is smaller than a preset probability threshold, if the prediction probability value corresponding to each window is smaller than the preset probability threshold, judging that the current video image to be detected does not contain the wheel of the annular cooler trolley, and determining the time information corresponding to the current video image to be detected as the target time.
That is, if the predicted probability value corresponding to each window in the current target detection area is smaller than the preset probability threshold, the current target detection area does not include the wheel of the circular cooler trolley, and it is determined that the current video image to be detected does not include the wheel of the circular cooler trolley.
Step S35, if the duration that the prediction probability value corresponding to the video image to be detected is smaller than the preset probability threshold value is longer than a preset time threshold value, determining that the wheel drop fault occurs in the circular cooler trolley; wherein the duration takes the target time as an initial time.
In a specific implementation manner, in this embodiment, wheel detection is performed through a target region to be detected, which is larger than the wheel region but smaller in range than the wheel region, and therefore may be considered as equal to the wheel area, so that it is determined that the preset time threshold is tThreshold value 2∈(Δtmin,Δtmax)；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
It can be understood that, after the detection area is reduced, and after one wheel is detected, the time for the next wheel to enter the target area is S/v, then it is determined that the time threshold for the wheel to fall off is larger than S/v, and meanwhile, obviously, the time for the next wheel to enter the target area is smaller than 2S/v, so that the value of the threshold is only required to be selected to be an appropriate value within the interval of S/v and 2S/v.
Referring to fig. 10, the present application discloses a wheel detecting apparatus for a trolley of a circular cooler, comprising,
a target data acquisition module 11; the method comprises the steps of acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler;
the video image detection module 12 is configured to, when any frame of the video image to be detected is acquired, input the video image to be detected to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain the wheels of the trolley of the circular cooler and corresponding label information;
the detection result obtaining module 13 is configured to obtain a predicted probability value that the current video image to be detected output by the trained classifier includes the wheels of the trolley of the circular cooler;
the first condition judgment module 14 is configured to judge whether the predicted probability value is smaller than a preset probability threshold, and if the predicted probability value is smaller than the preset probability threshold, determine that the current video image to be detected does not include wheels of the trolley of the circular cooler;
the target time determining module 15 is configured to determine the time information corresponding to the current video image to be detected as target time when the first condition determining module determines that the current video image to be detected does not include the wheels of the trolley of the circular cooler;
the second condition judgment module 16 is configured to judge that a wheel drop fault occurs in the circular cooler trolley if the duration that the prediction probability value corresponding to the obtained video image to be detected is smaller than the preset probability threshold is larger than a preset time threshold; wherein the duration takes the target time as an initial time.
Therefore, the method and the device for detecting the video images acquire the video images to be detected and the time information corresponding to each frame of the video images to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler, and when any frame of the video image to be detected is acquired, the video image to be detected is input to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain wheels of the circular cooler trolley and corresponding label information, then a prediction probability value of the wheels of the circular cooler trolley contained in a current video image to be detected output by the classifier after training is obtained, then whether the prediction probability value is smaller than a preset probability threshold value or not is judged, if the prediction probability value is smaller than the preset probability threshold value, the current video image to be detected is judged not to contain the wheels of the circular cooler trolley, time information corresponding to the current video image to be detected is determined as target time, and if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold value, is larger than the preset time threshold value, wheel dropping failure of the circular cooler trolley is judged; wherein the duration takes the target time as an initial time. Therefore, the video images acquired in real time aiming at the wheels of the circular cooler trolley are used for detection, and whether the wheel falling fault occurs in the circular cooler trolley can be detected timely, so that the safety accident is avoided.
The wheel detection device of the circular cooler trolley further comprises a preset time threshold value determination module, and the preset time threshold value determination module is used for determining the preset time threshold value by using the wheel distance between adjacent circular cooler trolleys.
In one specific embodiment, the video image is examinedThe detection module 12 is specifically configured to traverse the video image to be detected by using a sliding window, and sequentially input the video image to be detected in the sliding window to a pre-acquired classifier after training, so that the classifier after training completes image detection on the video image to be detected. Correspondingly, the preset time threshold determination module is specifically configured to determine that the preset time threshold is tThreshold value 1∈(0，Δtmin]；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
In another specific embodiment, the video image detection module 12 is specifically configured to intercept a target region to be detected from the video image to be detected; the area of the target area to be detected is larger than the wheel area of the wheel of the trolley of the circular cooler; traversing the target region to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected. Correspondingly, the preset time threshold determination module is specifically configured to,
determining the preset time threshold value as tThreshold value 2∈(Δtmin,Δtmax)；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
The wheel detection device of the circular cooler trolley further comprises a classifier training module for performing feature extraction on the target training sample by using a preset feature extraction algorithm; and training a preset classifier by using the target training sample after feature extraction to obtain the trained classifier.
For example, referring to fig. 11, fig. 11 is a schematic structural diagram of a wheel detection system of a circular cooler trolley disclosed in an embodiment of the present application, where the system includes a video camera, a camera light source, a host, a loudspeaker, and a terminal. The host in fig. 11 is used as the wheel detection device of the circular cooler trolley, and is used for processing images, the terminal is used for displaying alarm information and corresponding images, and the loudspeaker is used for giving out alarm prompt tones.
Referring to fig. 12, the embodiment of the present application discloses a wheel detection device for a loop cooling machine trolley, which includes a processor 21 and a memory 22; wherein, the memory 22 is used for saving computer programs; the processor 21 is configured to execute the computer program to implement the method for detecting the wheel of the circular cooler trolley disclosed in the foregoing embodiment.
For the specific process of the wheel detection method of the circular cooler trolley, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
Referring to fig. 13, the present application discloses a server 20 including a processor 21 and a memory 22 disclosed in the foregoing embodiments. For the steps that the processor 21 can specifically execute, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not described herein again.
Further, the server 20 in this embodiment may further specifically include a power supply 23, a communication interface 24, an input/output interface 25, and a communication bus 26; the power supply 23 is configured to provide operating voltage for each hardware device on the server 20; the communication interface 24 can create a data transmission channel between the server 20 and an external device, and a communication protocol followed by the communication interface is any communication protocol applicable to the technical solution of the present application, and is not specifically limited herein; the input/output interface 25 is configured to obtain external input data or output data to the outside, and a specific interface type thereof may be selected according to specific application requirements, which is not specifically limited herein.
Further, the embodiment of the present application also discloses a computer readable storage medium for storing a computer program, wherein the computer program is executed by a processor to implement the method for detecting the wheel of the loop cooling machine trolley disclosed in the foregoing embodiment.
For the specific process of the wheel detection method of the circular cooler trolley, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The method, the device, the equipment and the medium for detecting the wheel of the circular cooler trolley provided by the application are introduced in detail, specific examples are applied in the description to explain the principle and the implementation mode of the application, and the description of the above embodiments is only used for helping to understand the method and the core idea of the application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (10)
1. The method for detecting the wheels of the circular cooler trolley is characterized by comprising the following steps of:
acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler;
when any frame of the video image to be detected is acquired, inputting the video image to be detected to a pre-acquired post-training classifier; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain the wheels of the trolley of the circular cooler and corresponding label information;
acquiring the predicted probability value of the current video image to be detected output by the trained classifier, wherein the current video image to be detected contains the wheels of the trolley of the circular cooler;
judging whether the prediction probability value is smaller than a preset probability threshold value, if so, judging that the current video image to be detected does not contain wheels of the trolley of the circular cooler, and determining the time information corresponding to the current video image to be detected as target time;
if the duration of the obtained prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold, determining that the circular cooler trolley has a wheel drop fault; wherein the duration takes the target time as an initial time.
2. The method for detecting the wheels of the circular cooler trolley according to claim 1, wherein the inputting the video image to be detected into a pre-acquired trained classifier comprises:
traversing the video image to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected.
3. The method for detecting the wheels of the circular cooler trolley according to claim 1, wherein the inputting the video image to be detected into a pre-acquired trained classifier comprises:
intercepting a target to-be-detected area from the to-be-detected video image; the area of the target area to be detected is larger than the wheel area of the wheel of the trolley of the circular cooler;
traversing the target region to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected.
4. The method for detecting wheels of a trolley for a circular cooler according to claim 1, further comprising:
and determining the preset time threshold value by using the wheel distance of the adjacent circular cooler trolleys.
5. The method for detecting the wheels of the circular cooler trolley according to claim 4, wherein the step of determining the preset time threshold value by using the wheel distance between the adjacent circular cooler trolleys comprises the following steps:
if the video image to be detected is input to the pre-acquired classifier after training, including,
traversing the video image to be detected by using a sliding window, and sequentially inputting the video image to be detected in the sliding window to a pre-acquired classifier after training so that the classifier after training can complete image detection on the video image to be detected,
determining the preset time threshold value as tThreshold value 1∈(0，Δtmin]；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
6. The method for detecting the wheels of the circular cooler trolley according to claim 4, wherein the step of determining the preset time threshold value by using the wheel distance between the adjacent circular cooler trolleys comprises the following steps:
if the video image to be detected is input into a pre-acquired classifier after training, intercepting a target region to be detected from the video image to be detected; the target area to be detected is larger than the wheel area of the wheels of the trolley of the circular cooler, the target area to be detected is traversed by utilizing a sliding window, the video images to be detected in the sliding window are sequentially input to a pre-acquired classifier after training, so that the classifier after training finishes image detection aiming at the video images to be detected,
determining the preset time threshold value as tThreshold value 2∈(Δtmin,Δtmax)；
Wherein the content of the first and second substances,
s is the wheel distance between two adjacent trolleys of the ring cooling machine, and v is the running average speed of the trolleys of the ring cooling machine.
7. The method for detecting wheels of a ring cooler trolley according to any one of claims 1 to 6, further comprising:
extracting the features of the target training sample by using a preset feature extraction algorithm;
and training a preset classifier by using the target training sample after feature extraction to obtain the trained classifier.
8. The utility model provides a cold board car wheel detection device of ring which characterized in that includes:
a target data acquisition module; the method comprises the steps of acquiring a video image to be detected and time information corresponding to each frame of the video image to be detected; the video image to be detected is an image which is acquired by a camera in real time aiming at the wheels of the trolley of the circular cooler;
the video image detection module is used for inputting the video image to be detected to a pre-acquired post-training classifier when any frame of the video image to be detected is acquired; the trained classifier is obtained by training a preset classifier by using a target training sample; the target training samples comprise a first training sample and a second training sample; the first training sample comprises an image containing the wheels of the trolley of the circular cooler and corresponding label information; the second training sample is an image which does not contain the wheels of the trolley of the circular cooler and corresponding label information;
the detection result acquisition module is used for acquiring the prediction probability value of the current video image to be detected output by the trained classifier, which contains the wheels of the trolley of the circular cooler;
the first condition judgment module is used for judging whether the prediction probability value is smaller than a preset probability threshold value or not, and if the prediction probability value is smaller than the preset probability threshold value, judging that the current video image to be detected does not contain wheels of the trolley of the circular cooler;
the target time determining module is used for determining the time information corresponding to the current video image to be detected as target time under the condition that the first condition judging module judges that the current video image to be detected does not contain the wheels of the trolley of the circular cooler;
the second condition judgment module is used for judging that the wheel drop fault occurs in the circular cooler trolley if the duration of the acquired prediction probability value corresponding to the video image to be detected, which is smaller than the preset probability threshold, is larger than a preset time threshold; wherein the duration takes the target time as an initial time.
9. The wheel detection equipment of the circular cooler trolley is characterized by comprising a processor and a memory; wherein the content of the first and second substances,
the memory is used for storing a computer program;
the processor is used for executing the computer program to realize the circular cooler trolley wheel detection method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the method for cold ring trolley wheel detection according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010223896.5A CN112393617B (en) | 2020-03-26 | 2020-03-26 | Method, device, equipment and medium for detecting wheels of trolley of annular cooler |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010223896.5A CN112393617B (en) | 2020-03-26 | 2020-03-26 | Method, device, equipment and medium for detecting wheels of trolley of annular cooler |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112393617A true CN112393617A (en) | 2021-02-23 |
| CN112393617B CN112393617B (en) | 2022-06-03 |
Family
ID=74603815
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010223896.5A Active CN112393617B (en) | 2020-03-26 | 2020-03-26 | Method, device, equipment and medium for detecting wheels of trolley of annular cooler |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112393617B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114530043A (en) * | 2022-03-03 | 2022-05-24 | 上海闪马智能科技有限公司 | Event detection method and device, storage medium and electronic device |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006298208A (en) * | 2005-04-21 | 2006-11-02 | Aisin Aw Co Ltd | Parking support method and parking support device |
| US20080119987A1 (en) * | 2006-11-22 | 2008-05-22 | Thorsten Wilhelm Hestermeyer | Steering angle detection by means of ESC and EPAS |
| CN202853376U (en) * | 2012-08-17 | 2013-04-03 | 济钢集团有限公司 | Automatic detection device for trolley wheels |
| CN203980900U (en) * | 2014-08-15 | 2014-12-03 | 唐山市协力胶带运输设备有限公司 | The ring burning equipment running wheel detection system that comes off |
| CN105067287A (en) * | 2015-07-23 | 2015-11-18 | 北京京天威科技发展有限公司 | Motor train unit running fault dynamic image detecting system and method |
| CN105115301A (en) * | 2015-09-16 | 2015-12-02 | 广东韶钢松山股份有限公司 | Warning system for wheel disengagement of sintering circle-cooling machine |
| CN105501249A (en) * | 2014-10-10 | 2016-04-20 | 铁路发展服务公司 | System and method for detecting wheel condition |
| CN106766977A (en) * | 2016-12-20 | 2017-05-31 | 中冶华天南京工程技术有限公司 | Central cooler sideslip method of real-time |
| CN107273802A (en) * | 2017-05-16 | 2017-10-20 | 武汉华目信息技术有限责任公司 | A kind of detection method and device of railroad train brake shoe drill ring failure |
| CN207407690U (en) * | 2017-10-10 | 2018-05-25 | 中天钢铁集团有限公司 | Wheel of sintering trolley comes off early warning detection device |
| CN108896308A (en) * | 2018-07-02 | 2018-11-27 | 昆明理工大学 | A kind of wheel set bearing method for diagnosing faults based on probability envelope |
| CN109029377A (en) * | 2018-07-16 | 2018-12-18 | 银河水滴科技（北京）有限公司 | It is a kind of using visual analysis to detection car weight positioning square law device and system |
| CN109238756A (en) * | 2018-10-15 | 2019-01-18 | 哈尔滨市科佳通用机电股份有限公司 | Lorry operation troubles motion image detection equipment and detection method |
| CN109949269A (en) * | 2019-01-24 | 2019-06-28 | 武汉华目信息技术有限责任公司 | A kind of detection method and device of railroad train dust cap breakage failure |
| CN109975040A (en) * | 2017-12-28 | 2019-07-05 | 中冶长天国际工程有限责任公司 | A kind of pallet wheel detection method, apparatus and system |
| CN110793451A (en) * | 2019-10-31 | 2020-02-14 | 鞍山市海汇自动化有限公司 | Sintering trolley walking monitoring method and system based on laser ranging and image processing |
-
2020
- 2020-03-26 CN CN202010223896.5A patent/CN112393617B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006298208A (en) * | 2005-04-21 | 2006-11-02 | Aisin Aw Co Ltd | Parking support method and parking support device |
| US20080119987A1 (en) * | 2006-11-22 | 2008-05-22 | Thorsten Wilhelm Hestermeyer | Steering angle detection by means of ESC and EPAS |
| CN202853376U (en) * | 2012-08-17 | 2013-04-03 | 济钢集团有限公司 | Automatic detection device for trolley wheels |
| CN203980900U (en) * | 2014-08-15 | 2014-12-03 | 唐山市协力胶带运输设备有限公司 | The ring burning equipment running wheel detection system that comes off |
| CN105501249A (en) * | 2014-10-10 | 2016-04-20 | 铁路发展服务公司 | System and method for detecting wheel condition |
| CN105067287A (en) * | 2015-07-23 | 2015-11-18 | 北京京天威科技发展有限公司 | Motor train unit running fault dynamic image detecting system and method |
| CN105115301A (en) * | 2015-09-16 | 2015-12-02 | 广东韶钢松山股份有限公司 | Warning system for wheel disengagement of sintering circle-cooling machine |
| CN106766977A (en) * | 2016-12-20 | 2017-05-31 | 中冶华天南京工程技术有限公司 | Central cooler sideslip method of real-time |
| CN107273802A (en) * | 2017-05-16 | 2017-10-20 | 武汉华目信息技术有限责任公司 | A kind of detection method and device of railroad train brake shoe drill ring failure |
| CN207407690U (en) * | 2017-10-10 | 2018-05-25 | 中天钢铁集团有限公司 | Wheel of sintering trolley comes off early warning detection device |
| CN109975040A (en) * | 2017-12-28 | 2019-07-05 | 中冶长天国际工程有限责任公司 | A kind of pallet wheel detection method, apparatus and system |
| CN108896308A (en) * | 2018-07-02 | 2018-11-27 | 昆明理工大学 | A kind of wheel set bearing method for diagnosing faults based on probability envelope |
| CN109029377A (en) * | 2018-07-16 | 2018-12-18 | 银河水滴科技（北京）有限公司 | It is a kind of using visual analysis to detection car weight positioning square law device and system |
| CN109238756A (en) * | 2018-10-15 | 2019-01-18 | 哈尔滨市科佳通用机电股份有限公司 | Lorry operation troubles motion image detection equipment and detection method |
| CN109949269A (en) * | 2019-01-24 | 2019-06-28 | 武汉华目信息技术有限责任公司 | A kind of detection method and device of railroad train dust cap breakage failure |
| CN110793451A (en) * | 2019-10-31 | 2020-02-14 | 鞍山市海汇自动化有限公司 | Sintering trolley walking monitoring method and system based on laser ranging and image processing |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114530043A (en) * | 2022-03-03 | 2022-05-24 | 上海闪马智能科技有限公司 | Event detection method and device, storage medium and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112393617B (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bertini et al. | Multi-scale and real-time non-parametric approach for anomaly detection and localization | |
| JP3892059B2 (en) | Moving body tracking device | |
| Cao et al. | Linear SVM classification using boosting HOG features for vehicle detection in low-altitude airborne videos | |
| CN111310645B (en) | Method, device, equipment and storage medium for warning overflow bin of goods accumulation | |
| US10970823B2 (en) | System and method for detecting motion anomalies in video | |
| CN104966304A (en) | Kalman filtering and nonparametric background model-based multi-target detection tracking method | |
| Pena-Gonzalez et al. | Computer vision based real-time vehicle tracking and classification system | |
| Aydin | A new approach based on firefly algorithm for vision-based railway overhead inspection system | |
| CN102902955A (en) | Method and system for intelligently analyzing vehicle behaviour | |
| US20200293794A1 (en) | Methods and systems for accurately recognizing vehicle license plates | |
| CN105631418A (en) | People counting method and device | |
| CN111310826B (en) | Method and device for detecting labeling abnormality of sample set and electronic equipment | |
| CN111079621B (en) | Method, device, electronic equipment and storage medium for detecting object | |
| US20220198801A1 (en) | Method and system for detecting fire and smoke | |
| CN112393617B (en) | Method, device, equipment and medium for detecting wheels of trolley of annular cooler | |
| CN111950566A (en) | Rotation-invariant HOG infrared image power equipment identification method | |
| CN103150547A (en) | Vehicle tracking method and device | |
| JP2021086379A (en) | Information processing apparatus, information processing method, program, and method of generating learning model | |
| CN112836683A (en) | License plate recognition method, device, equipment and medium for portable camera equipment | |
| Astawa et al. | Detection of license plate using sliding window, Histogram of oriented gradient, and support vector machines method | |
| CN112818753A (en) | Pit falling object detection method, device and system | |
| Hervieu et al. | A statistical video content recognition method using invariant features on object trajectories | |
| JP7125843B2 (en) | Fault detection system | |
| WO2023221770A1 (en) | Dynamic target analysis method and apparatus, device, and storage medium | |
| CN111862023A (en) | Pantograph structure abnormity detection method based on computer vision |
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 |