CN117734683A - Underground vehicle anti-collision safety early warning decision-making method - Google Patents
Underground vehicle anti-collision safety early warning decision-making method Download PDFInfo
- Publication number
- CN117734683A CN117734683A CN202410182953.8A CN202410182953A CN117734683A CN 117734683 A CN117734683 A CN 117734683A CN 202410182953 A CN202410182953 A CN 202410182953A CN 117734683 A CN117734683 A CN 117734683A
- Authority
- CN
- China
- Prior art keywords
- alarm
- collision
- vehicle
- information
- decision
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 11
- 230000003287 optical effect Effects 0.000 claims description 9
- 230000001629 suppression Effects 0.000 claims description 5
- 239000003245 coal Substances 0.000 claims description 4
- 238000005286 illumination Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 239000000428 dust Substances 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 150000003839 salts Chemical class 0.000 claims description 3
- 230000036541 health Effects 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000004888 barrier function Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention provides an underground vehicle anti-collision safety early warning decision-making method, which comprises the following steps: collecting front road information, inputting the front road information into a target detection network to obtain characteristic information, generating bounding boxes with corresponding dimensions according to the characteristic information, and positioning the position information of all targets; determining bounding boxes with associated characteristic information as the same target under different time sequences in the road information, acquiring relative positions between the target and the vehicle-mounted sensor under continuous time sequences, and calculating relative movement speed and track of the target; according to the relative movement speed and the track, the collision time between the current vehicle-mounted sensor and all dynamic targets in the running direction of the vehicle is calculated, and the collision time is corrected based on an alarm decision algorithm and is sent to a user side. The invention has the beneficial effects that: dynamic barriers can be effectively identified in underground narrow roadways, the occurrence rate of collision accidents of intelligent mines is effectively reduced, and the life health of personnel and the safety of equipment property in vehicle running are ensured.
Description
Technical Field
The invention belongs to the field of automatic driving, and particularly relates to an underground vehicle anti-collision safety early warning decision method.
Background
Due to dim underground light, low identification degree, noisy environment and the like, vehicle transportation is always a disaster area of mine accidents, and the vehicle transportation once accounts for more than 20% of the total accident amount, so that a large amount of equipment property damage and life health disability are caused, the safe and green operation of an intelligent mine auxiliary transportation system is seriously balanced, and the hazard is self-evident.
According to European Union statistics, a vehicle safety early warning system is a key for solving the problem, road conditions are continuously detected through a vehicle-mounted high-precision sensor, upcoming collision hidden dangers are predicted, a driver is reminded of avoiding risks by combining an intelligent decision-making method, the collision incidence rate and the casualty rate of the driver of vehicles can be effectively reduced by more than 50%, and the system is moved to a down-hole scene to be an important discussion issue of each mine enterprise.
However, most of the existing vehicle safety early warning method researches are built in a normal road environment, are not based on underground severe environments, and cannot adapt to severe noise, dim light, narrow roadways and the like in a coal mine environment. Therefore, the vehicle anti-collision safety early warning decision-making method which can be suitable for underground environment, is strong and robust and is high in accuracy is provided, and is a key problem to be solved in the mine intelligent transportation system.
Disclosure of Invention
In view of the above, the present invention aims to provide a method for deciding anti-collision safety precaution of a downhole vehicle, so as to solve at least one of the above technical problems.
In order to achieve the above purpose, the technical scheme of the invention is realized as follows:
the first aspect of the invention provides a method for deciding anti-collision safety pre-warning of a downhole vehicle, which comprises the following steps:
the vehicle-mounted sensor continuously collects front road information, inputs the front road information into the target detection network after preprocessing the front road information to obtain characteristic information, generates a bounding box with corresponding dimension according to the characteristic information, and locates the position information of all targets;
determining bounding boxes with associated characteristic information as the same target under different time sequences in the road information, acquiring relative positions between the target and the vehicle-mounted sensor under continuous time sequences, and calculating relative movement speed and track of the target;
according to the relative movement speed and the track, the collision time between the current vehicle-mounted sensor and all dynamic targets in the running direction of the vehicle is calculated, and the collision time is corrected based on an alarm decision algorithm and is sent to a user side.
Further, the training process before the application of the target detection network comprises the following steps: and inputting historical road information to train the target detection network once, and inputting the characteristics of the obstacle in the narrow roadway into the trained target detection network once to train adaptively.
Further, adding the interference information into the historical road information, and performing secondary training on the target detection network after the adaptive training; the interference information comprises electric noise, illumination difference, dust fog interference, spiced salt noise, rayleigh noise and Gaussian mixture noise of a coal mine site.
Further, the vehicle-mounted sensor includes a radar sensor and an optical sensor;
the radar sensor acquires reflected waves in a front road, identifies the positions of reflection points with characteristic information, and outputs a three-dimensional bounding box;
the optical sensor acquires a real-time image in a front road, identifies the position of a pixel point with characteristic information in an image, and outputs a two-dimensional bounding box;
the optical sensor is selected based on the downhole low-illuminance dim light environment adaptability.
Further, the collision time is based on the time of collision between the current vehicle and the target under the condition that the relative speed between the current vehicle and the target on the same path is unchanged;
and according to the changes of the relative motion speed and the track, the collision time is modified in real time and sent to the client in real time.
Further, the target detection network performs deep feature extraction operation on the preprocessed front road information, abstracts and fits the extracted feature information, outputs candidate bounding boxes and position information of a plurality of targets, performs non-maximum suppression on the candidate bounding boxes, and screens out targets with high confidence.
Further, an alarm threshold value is preset in the alarm decision algorithm, and an alarm is given out by the alarm when the collision time is smaller than the alarm threshold value;
the alarm comprises an audible alarm, a lamplight alarm and a vibration alarm, and a plurality of alarms with different alarm forms are combined for three-dimensional alarm;
the alarm sound of the sound alarm is higher than 60 dB, and the frequency is 300-5000 Hz.
Further, the process of sending out the alarm by the alarm comprises the following steps:
triggering the pre-alarm signal to be converted into a pre-alarm state when the collision time is close to the alarm threshold value, and continuously recording the current collision time by the alarm;
if the collision time is smaller than the alarm threshold value in the pre-alarm state, the alarm state is changed to be continuously activated;
otherwise, counting the duration of the pre-alarm state, and switching to the normal state after the duration exceeds the pre-alarm maintaining time.
A second aspect of the present invention provides an electronic device comprising a processor and a memory communicatively coupled to the processor for storing instructions executable by the processor, characterized by: the processor is configured to execute the method for determining anti-collision safety precaution of a downhole vehicle according to any one of the first aspect.
A third aspect of the present invention provides a server, characterized in that: the system comprises at least one processor and a memory communicatively connected with the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to cause the at least one processor to perform a downhole vehicle anti-collision safety pre-warning decision method according to any one of the first aspects.
Compared with the prior art, the underground vehicle anti-collision safety early warning decision method has the following beneficial effects:
the vehicle-mounted sensor is used for collecting data in real time, dynamic obstacles can be effectively identified in underground narrow roadways by means of the robustness, reliability and stability of an intelligent AI algorithm, an alarm signal is output in real time to warn a driver to avoid collision, the occurrence rate of collision accidents of an intelligent mine is effectively reduced, and the life health and equipment property safety of personnel in vehicle driving are guaranteed.
The hardware equipment only comprises the vehicle-mounted sensor, the alarm and the arithmetic unit, so that the hardware composition is simple and clear, the equipment cost is greatly reduced, and the maintainability of the system is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention. In the drawings:
fig. 1 is a schematic workflow diagram of a method for making an anti-collision safety warning decision for an underground vehicle according to an embodiment of the present invention.
Detailed Description
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
The invention will be described in detail below with reference to the drawings in connection with embodiments.
Embodiment one:
as shown in fig. 1, a method for deciding anti-collision safety pre-warning of a downhole vehicle includes:
s1, continuously collecting front road information by a vehicle-mounted sensor, preprocessing the front road information, inputting the front road information into a target detection network to obtain characteristic information, generating bounding boxes with corresponding dimensions according to the characteristic information, and positioning the position information of all targets;
s2, judging surrounding frames with associated characteristic information as the same target under different time sequences in the road information, acquiring relative positions between the target and the vehicle-mounted sensor under continuous time sequences, and calculating the relative movement speed and track of the target;
s3, according to the relative movement speed and the track, calculating the collision time between the current vehicle-mounted sensor and all dynamic targets in the running direction of the vehicle, correcting the collision time based on an alarm decision algorithm, and sending the corrected collision time to the user side.
The training process before the application of the target detection network comprises the following steps: and inputting historical road information to train the target detection network once, and inputting the characteristics of the obstacle in the narrow roadway into the trained target detection network once to train adaptively.
The specific process comprises the following steps:
data collection and preparation: collecting a data set containing historical road information and obstacle characteristics in a narrow roadway, wherein the data set comprises image or video data and corresponding labeling information, such as bounding boxes and class labels, and performing image enhancement, data enhancement and label conversion operation on the data to improve the quality and diversity of the data;
pre-training: pre-training a target detection network by using a large-scale general data set, and inputting historical road information into the network in the pre-training process;
fine tuning: the characteristics of the obstacles in the narrow roadway are input into the pre-trained target detection network for adjustment, the fine adjustment process uses a smaller learning rate so as to keep the general characteristics learned by the network in the pre-training stage, and the characteristics of the obstacles in the narrow roadway are focused more, so that the fine adjustment aims at enabling the network to adapt to the target detection task in the narrow roadway scene.
Adding the interference information into the historical road information, and performing secondary training on the target detection network after the adaptive training; the interference information comprises electric noise, illumination difference, dust fog interference, spiced salt noise, rayleigh noise and Gaussian mixture noise of a coal mine site.
The target detection network is trained by adding the interference information to the historical road information, so that the target detection network has better anti-interference performance, and the specific process is as follows:
adding interference information into historical road information by overlapping interference pixels, adjusting illumination conditions and adding noise to generate a training sample with interference;
dividing the training samples with interference into small batches for processing, wherein each batch comprises a certain number of training samples and corresponding labels;
inputting a batch of training samples into a target detection network for forward propagation, wherein the network calculates a prediction result according to the current weight parameters, wherein the prediction result comprises the category and the position of the target;
comparing the predicted result of the network with a real label, calculating a loss function, calculating a gradient through a back propagation algorithm according to the value of the loss function, and updating the weight parameter of the network;
a random gradient descent algorithm is used for adjusting the value of the weight parameter according to the direction and the size of the gradient and the calculated gradient so as to gradually reduce the loss function;
and repeatedly executing the steps until all training samples are traversed.
The vehicle-mounted sensor comprises a radar sensor and an optical sensor;
the radar sensor acquires reflected waves in a front road, identifies the positions of reflection points with characteristic information, and outputs a three-dimensional bounding box;
the optical sensor acquires a real-time image in a front road, identifies the position of a pixel point with characteristic information in an image, and outputs a two-dimensional bounding box;
the optical sensor is selected based on the underground low-illuminance dim light environment adaptability, and particularly an explosion-proof low-illuminance camera or an infrared imaging camera.
The collision time is based on the time of collision between the current vehicle and the target under the condition that the relative speed between the current vehicle and the target on the same path is unchanged; and according to the changes of the relative motion speed and the track, the collision time is modified in real time and sent to the client in real time.
The target detection network performs deep feature extraction operation on the preprocessed front road information, abstracts and fits the extracted feature information, outputs candidate bounding boxes and position information of a plurality of targets, performs non-maximum suppression on the candidate bounding boxes, and screens out targets with high confidence.
The specific process is as follows:
extracting features of the preprocessed road information through the first few layers of convolution layers of the target detection network, and learning feature representations of different scales and abstraction levels by the convolution layers to obtain low-level edge and texture features and high-level semantic features;
after the feature is extracted, the target detection network performs abstraction and fitting on the extracted feature information through further operations such as convolution, pooling and full connection layer, and gradually converts low-level features into higher-level semantic features so as to better represent the shape, texture and context information of the target;
after feature abstraction and fitting, the target detection network generates candidate bounding boxes and position information of a plurality of targets through convolution operation and a full connection layer;
and screening the candidate frames by using a non-maximum suppression algorithm, sorting and screening according to the overlapping degree and the confidence coefficient between the candidate frames, selecting the candidate frames which have the highest confidence coefficient and are not overlapped as a final detection result, and outputting the surrounding frame and the position information of the high-confidence coefficient target after the non-maximum suppression screening.
An alarm threshold value is preset in the alarm decision algorithm, and an alarm is given out by the alarm when the collision time is smaller than the alarm threshold value;
the alarm comprises an audible alarm, a lamplight alarm and a vibration alarm, and a plurality of alarms with different alarm forms are combined for three-dimensional alarm;
the alarm sound of the sound alarm is higher than 60 dB, and the frequency is 300-5000 Hz.
The alarm sending process comprises the following steps:
triggering the pre-alarm signal to be converted into a pre-alarm state when the collision time is close to the alarm threshold value, and continuously recording the current collision time by the alarm;
if the collision time is smaller than the alarm threshold value in the pre-alarm state, the alarm state is changed to be continuously activated;
otherwise, counting the duration of the pre-alarm state, and switching to the normal state after the duration exceeds the pre-alarm maintaining time.
Embodiment two:
an electronic device comprising a processor and a memory communicatively coupled to the processor for storing processor-executable instructions, characterized in that: the processor is configured to execute the method for determining anti-collision safety precaution of a downhole vehicle according to any one of the above embodiments.
Embodiment III:
a server, characterized by: the method of determining a safety warning of a vehicle downhole according to any one of the embodiments, comprising at least one processor and a memory communicatively coupled to the processor, the memory storing instructions executable by the at least one processor to cause the at least one processor to perform a method of determining a safety warning of a vehicle downhole.
Those of ordinary skill in the art will appreciate that the elements and method steps of each example described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the elements and steps of each example have been described generally in terms of functionality in the foregoing description to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided in this application, it should be understood that the disclosed methods and systems may be implemented in other ways. For example, the above-described division of units is merely a logical function division, and there may be another division manner when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted or not performed. The units may or may not be physically separate, and components shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiment of the present invention.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention, and are intended to be included within the scope of the appended claims and description.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. The underground vehicle anti-collision safety early warning decision-making method is characterized by comprising the following steps of:
the vehicle-mounted sensor continuously collects front road information, inputs the front road information into the target detection network after preprocessing the front road information to obtain characteristic information, generates a bounding box with corresponding dimension according to the characteristic information, and locates the position information of all targets;
determining bounding boxes with associated characteristic information as the same target under different time sequences in the road information, acquiring relative positions between the target and the vehicle-mounted sensor under continuous time sequences, and calculating relative movement speed and track of the target;
according to the relative movement speed and the track, the collision time between the current vehicle-mounted sensor and all dynamic targets in the running direction of the vehicle is calculated, and the collision time is corrected based on an alarm decision algorithm and is sent to a user side.
2. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 1, wherein the method comprises the steps of:
the training process before the application of the target detection network comprises the following steps: and inputting historical road information to train the target detection network once, and inputting the characteristics of the obstacle in the narrow roadway into the trained target detection network once to train adaptively.
3. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 2, wherein the method comprises the steps of:
adding the interference information into the historical road information, and performing secondary training on the target detection network after the adaptive training; the interference information comprises electric noise, illumination difference, dust fog interference, spiced salt noise, rayleigh noise and Gaussian mixture noise of a coal mine site.
4. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 1, wherein the method comprises the steps of:
the vehicle-mounted sensor comprises a radar sensor and an optical sensor;
the radar sensor acquires reflected waves in a front road, identifies the positions of reflection points with characteristic information, and outputs a three-dimensional bounding box;
the optical sensor acquires a real-time image in a front road, identifies the position of a pixel point with characteristic information in an image, and outputs a two-dimensional bounding box;
the optical sensor is selected based on the downhole low-illuminance dim light environment adaptability.
5. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 1, wherein the method comprises the steps of:
the collision time is based on the time of collision between the current vehicle and the target under the condition that the relative speed between the current vehicle and the target on the same path is unchanged;
and according to the changes of the relative motion speed and the track, the collision time is modified in real time and sent to the client in real time.
6. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 1, wherein the method comprises the steps of:
the target detection network performs deep feature extraction operation on the preprocessed front road information, abstracts and fits the extracted feature information, outputs candidate bounding boxes and position information of a plurality of targets, performs non-maximum suppression on the candidate bounding boxes, and screens out targets with high confidence.
7. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 1, wherein the method comprises the steps of:
an alarm threshold value is preset in the alarm decision algorithm, and an alarm is given out by the alarm when the collision time is smaller than the alarm threshold value;
the alarm comprises an audible alarm, a lamplight alarm and a vibration alarm, and a plurality of alarms with different alarm forms are combined for three-dimensional alarm;
the alarm sound of the sound alarm is higher than 60 dB, and the frequency is 300-5000 Hz.
8. The method for making a decision for anti-collision safety precaution for a downhole vehicle according to claim 7, wherein the method comprises the steps of:
the alarm sending process comprises the following steps:
triggering the pre-alarm signal to be converted into a pre-alarm state when the collision time is close to the alarm threshold value, and continuously recording the current collision time by the alarm;
if the collision time is smaller than the alarm threshold value in the pre-alarm state, the alarm state is changed to be continuously activated;
otherwise, counting the duration of the pre-alarm state, and switching to the normal state after the duration exceeds the pre-alarm maintaining time.
9. An electronic device comprising a processor and a memory communicatively coupled to the processor for storing processor-executable instructions, characterized in that: the processor is configured to perform a method for determining collision avoidance safety precautions for a downhole vehicle according to any of claims 1-8.
10. A server, characterized by: comprising at least one processor and a memory communicatively coupled to the processor, the memory storing instructions executable by the at least one processor to cause the at least one processor to perform a downhole vehicle collision safety warning decision method as recited in any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410182953.8A CN117734683B (en) | 2024-02-19 | Underground vehicle anti-collision safety early warning decision-making method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410182953.8A CN117734683B (en) | 2024-02-19 | Underground vehicle anti-collision safety early warning decision-making method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117734683A true CN117734683A (en) | 2024-03-22 |
| CN117734683B CN117734683B (en) | 2024-05-24 |
Family
ID=
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060190124A1 (en) * | 2003-03-25 | 2006-08-24 | Maekelae Hannu | Arrangement for collision prevention of mine vehicle |
| CN107561552A (en) * | 2017-08-16 | 2018-01-09 | 北京矿冶研究总院 | Anti-collision method and device for underground mine trackless equipment |
| CN107972662A (en) * | 2017-10-16 | 2018-05-01 | 华南理工大学 | To anti-collision warning method before a kind of vehicle based on deep learning |
| CN113370977A (en) * | 2021-05-06 | 2021-09-10 | 上海大学 | Intelligent vehicle forward collision early warning method and system based on vision |
| CN115649158A (en) * | 2022-11-11 | 2023-01-31 | 中煤电气有限公司 | Mine vehicle anti-collision method, equipment and storage medium |
| CN115903779A (en) * | 2022-10-14 | 2023-04-04 | 平煤神马建工集团有限公司 | Intelligent early warning system and method for shield tunnel electric locomotive |
| CN117087543A (en) * | 2023-07-25 | 2023-11-21 | 湖北三江航天红峰控制有限公司 | Intelligent auxiliary driving system for mining operation vehicle |
| CN117351298A (en) * | 2023-09-06 | 2024-01-05 | 华能伊敏煤电有限责任公司 | Mine operation vehicle detection method and system based on deep learning |
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060190124A1 (en) * | 2003-03-25 | 2006-08-24 | Maekelae Hannu | Arrangement for collision prevention of mine vehicle |
| CN107561552A (en) * | 2017-08-16 | 2018-01-09 | 北京矿冶研究总院 | Anti-collision method and device for underground mine trackless equipment |
| CN107972662A (en) * | 2017-10-16 | 2018-05-01 | 华南理工大学 | To anti-collision warning method before a kind of vehicle based on deep learning |
| CN113370977A (en) * | 2021-05-06 | 2021-09-10 | 上海大学 | Intelligent vehicle forward collision early warning method and system based on vision |
| CN115903779A (en) * | 2022-10-14 | 2023-04-04 | 平煤神马建工集团有限公司 | Intelligent early warning system and method for shield tunnel electric locomotive |
| CN115649158A (en) * | 2022-11-11 | 2023-01-31 | 中煤电气有限公司 | Mine vehicle anti-collision method, equipment and storage medium |
| CN117087543A (en) * | 2023-07-25 | 2023-11-21 | 湖北三江航天红峰控制有限公司 | Intelligent auxiliary driving system for mining operation vehicle |
| CN117351298A (en) * | 2023-09-06 | 2024-01-05 | 华能伊敏煤电有限责任公司 | Mine operation vehicle detection method and system based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109212520B (en) | Road condition sensing abnormal event detection alarm system and method for omnibearing detection radar | |
| CN108725440B (en) | Forward collision control method and apparatus, electronic device, program, and medium | |
| KR101095528B1 (en) | An outomatic sensing system for traffic accident and method thereof | |
| CN112951000B (en) | Large-scale vehicle blind area bidirectional early warning system | |
| Gandhi et al. | Pedestrian protection systems: Issues, survey, and challenges | |
| US11380105B2 (en) | Identification and classification of traffic conflicts | |
| EP1553516B1 (en) | Pedestrian extracting apparatus | |
| CN105809679A (en) | Mountain railway side slope rockfall detection method based on visual analysis | |
| CN109703460A (en) | The complex scene adaptive vehicle collision warning device and method for early warning of multi-cam | |
| CN111507237A (en) | Video-based abnormal event detection method and device applied to expressway | |
| CN111582130B (en) | Traffic behavior perception fusion system and method based on multi-source heterogeneous information | |
| KR20220146670A (en) | Traffic anomaly detection methods, devices, devices, storage media and programs | |
| CN115205796B (en) | Rail line foreign matter intrusion monitoring and risk early warning method and system | |
| CN115083088A (en) | Railway perimeter intrusion early warning method | |
| CN113484858A (en) | Intrusion detection method and system | |
| KR102423550B1 (en) | Method for determining traffic violation vehicle which exists on driving route of autonomous vehicle and the device using the same | |
| CN114972911A (en) | Method and equipment for collecting and processing output data of automatic driving perception algorithm model | |
| CN113792598B (en) | Vehicle-mounted camera-based vehicle collision prediction system and method | |
| CN117734683B (en) | Underground vehicle anti-collision safety early warning decision-making method | |
| CN105426852B (en) | Vehicle-mounted monocular LONG WAVE INFRARED camera pedestrian recognition method | |
| CN117734683A (en) | Underground vehicle anti-collision safety early warning decision-making method | |
| CN109278759A (en) | A kind of safe driving of vehicle auxiliary system | |
| CN117423157A (en) | Mine abnormal video action understanding method combining migration learning and regional invasion | |
| CN116563801A (en) | Traffic accident detection method, device, electronic equipment and medium | |
| KR20210023273A (en) | Apparatus and method for judging tunnel situation |
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 |