CN116434142A - Airport runway foreign matter detection method and system - Google Patents
Airport runway foreign matter detection method and system Download PDFInfo
- Publication number
- CN116434142A CN116434142A CN202310329815.3A CN202310329815A CN116434142A CN 116434142 A CN116434142 A CN 116434142A CN 202310329815 A CN202310329815 A CN 202310329815A CN 116434142 A CN116434142 A CN 116434142A
- Authority
- CN
- China
- Prior art keywords
- airport runway
- image
- foreign matter
- foreign
- runway
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 claims abstract description 38
- 230000000007 visual effect Effects 0.000 claims abstract description 27
- 230000002159 abnormal effect Effects 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 19
- 230000035945 sensitivity Effects 0.000 claims description 15
- 238000009826 distribution Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 230000033001 locomotion Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000003321 amplification Effects 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 claims description 2
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 8
- 238000005286 illumination Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000013145 classification model Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 230000010006 flight Effects 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 230000006641 stabilisation Effects 0.000 description 2
- 238000011105 stabilization Methods 0.000 description 2
- 239000004575 stone Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/147—Details of sensors, e.g. sensor lenses
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a method and a system for detecting foreign matters on an airport runway, which belong to the field of image processing, wherein the method comprises the following steps: acquiring an airport runway image; extracting visual features from the airport runway images by a feature extractor; processing visual features extracted from the airport runway images through a pre-trained foreign matter detection model to obtain an abnormal probability value of each position in the airport runway images; and identifying the foreign matters in the airport runway images through the abnormal probability values to obtain a foreign matter detection result. The invention can efficiently and accurately detect and identify the foreign matters on the airport runway.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a system for detecting foreign matters on an airport runway.
Background
At present, civil aviation in China is in a high-speed development period. According to the planning, by 2030, china comprehensively builds a safe and efficient modern civil aviation system, and realizes the transition from the civil aviation large country to the civil aviation strong country.
Continuous safety is a permanent topic of civil aviation development, and the problem of threat to safe operation of aircraft in the flight area is also receiving increasing attention. Runway foreign matter is the main hidden danger threatening civil aviation safety operation, and the runway foreign matter automatic monitoring system is the guarantee of aviation safety. It is reported that various unsafe events caused by runway foreign bodies take off and land up to 4 times per ten thousand times, and the loss caused by runway foreign bodies is at least between 30 and 40 hundred million dollars worldwide each year.
Runway foreign matter, i.e., foreign matter that can damage the aircraft, such as metal parts, waterproof plastic cloths, crushed stones, paper scraps, leaves, etc., is collectively referred to as foreign matter. Aircraft are quite fragile with respect to foreign objects, a small piece of plastic cloth sucked into the engine may cause a stop, a small screw or metal piece or even a sharp stone may puncture the tire and cause a burst, and the resulting tire fragments may in turn scratch the aircraft.
At present, the foreign matter detection methods of various airports are generally classified into the following methods:
1. manual inspection: the foreign objects on the runway are manually patrolled and inspected during the flight gap, including before the flight (before the first flight started every day), after the flight (after the last flight every day), in the air (flight gap) and after the aircraft forced landing, by the relevant personnel driving the vehicle into the airport runway. However, the method has the defects of high labor intensity, low detection efficiency, easiness in missed detection and the like, and meanwhile, the possibility of manual detection is smaller and smaller along with the increase of the flight density.
2. An edge light type photoelectric composite detection system: which is typically a combination of radar and video, the device is mounted on the road shoulder. However, the method has the problems of numerous equipment, large field construction, long time consumption, high cost, high maintenance difficulty and the like.
3. Conventional deep learning-based approaches: the method comprises the steps of obtaining a classification model by classifying, identifying and training the foreign matters in the airport runway, and detecting the foreign matters in the airport runway based on the classification model. However, the airport runway foreign matter detection needs to identify various target types, large to vehicles, small to screws and various fragments and parts, so that the mode of model training and identifying various foreign matters by stacking a large number of samples can cause model training to sink into infinite increase of foreign matter types and infinite update iteration of sample models, various foreign matters cannot be accurately identified in practical application, and obviously, the mode cannot meet practical requirements.
Disclosure of Invention
Aiming at the problems of high labor consumption, high equipment cost investment and low foreign matter identification precision in the airport runway foreign matter detection method in the prior art, the invention aims to provide the airport runway foreign matter detection method and system so as to at least partially solve the problems.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
in a first aspect, the present invention provides a method for detecting foreign matter on an airport runway, comprising the steps of:
acquiring an airport runway image;
extracting visual features from the airport runway images by a feature extractor;
processing visual features extracted from the airport runway images through a pre-trained foreign matter detection model to obtain an abnormal probability value of each position in the airport runway images;
and identifying the foreign matters in the airport runway images through the abnormal probability value to obtain a foreign matter detection result.
Preferably, the pre-trained foreign object detection model is obtained by:
visual feature extraction is carried out on the airport runway sample image without foreign matters through a feature extractor;
training a two-dimensional normalized flow implemented FastFlow algorithm using visual features extracted from the airport runway sample image, such that the two-dimensional normalized flow implemented FastFlow algorithm society converts the input visual features into processable distributions, and obtains an abnormal probability value for each position in the airport runway sample image in an inference stage, thereby obtaining the foreign object detection model.
Preferably, the step of acquiring the runway image includes:
firstly dividing an airport runway into a plurality of preset positions; and sequentially carrying out round-robin on a plurality of preset positions at preset scanning time by using an image acquisition device, so as to obtain airport runway images about each preset position.
Preferably, the image acquisition device is a tele laser cradle head.
Preferably, the image acquisition device and the airport runway are calibrated by calibration technology so as to determine the position of the foreign matter on the airport runway after the foreign matter is detected.
Preferably, the step of identifying the foreign matter in the airport runway image by the abnormal probability value includes:
and judging whether the abnormal probability value of each position in the airport runway image is higher than a sensitivity threshold value, and judging that the position is foreign matter.
Preferably, when the airport runway image is detected to contain an aircraft, the sensitivity threshold is adjusted by a preset amplitude reduction, so that the false alarm rate is reduced; wherein whether the airport runway image contains an aircraft is detected through an object detection algorithm.
Preferably, in the case where the airport runway image is detected to contain an aircraft within a preset period of time and the airport runway image is detected to not contain an aircraft, the sensitivity threshold is adjusted with a preset increase, thereby reducing the rate of false negatives.
Preferably, the foreign matter detection result is obtained by further comprising the steps of:
tracking the detected foreign matters through a track tracking algorithm, so as to obtain the motion state of the foreign matters, and filtering false alarms based on the principle of the foreign matters not moving.
In a second aspect, the present invention also provides an airport runway foreign matter detection system, comprising
The image acquisition module is used for receiving an airport runway image to be detected;
the feature extraction module is used for extracting visual features from the airport runway images;
the model calling module is used for calling a pre-trained foreign matter detection model to process the airport runway image and obtain an abnormal probability value of each position in the airport runway image;
and the foreign matter identification module is used for identifying the foreign matters in the airport runway images through the abnormal probability value to obtain a foreign matter detection result.
By adopting the technical scheme, the invention has the following beneficial effects: the invention adopts the FastFlow algorithm realized by the two-dimensional normalized flow and uses the FastFlow algorithm as a probability distribution estimator, and the FastFlow algorithm is used together with any depth feature extractors such as ResNet, a visual converter and the like for unsupervised anomaly detection and positioning. In the training stage, the FastFlow academy converts the input visual characteristics into processable distribution, and obtains an abnormal probability value in the reasoning stage; in actual application, the anomaly probability value can be calculated on the input airport runway image through FastFlow, and the foreign matters in the airport runway image can be judged and identified by combining the set sensitivity threshold. The method is not limited by the types of the foreign matters when the foreign matters are identified, and has extremely high detection rate in the actual detection of the foreign matters on the airport runway.
Drawings
Fig. 1 is a flowchart of a method for detecting foreign matters on an airport runway according to an embodiment of the invention.
Fig. 2 is a flowchart of adjusting a sensitivity threshold based on detecting an aircraft in a second embodiment of the invention.
Fig. 3 is a flowchart of a method for detecting foreign matters on an airport runway according to a third embodiment of the present invention.
Fig. 4 is a schematic structural diagram of an airport runway foreign matter detection system according to a fourth embodiment of the invention.
Fig. 5 is a schematic structural diagram of an electronic device in a fifth embodiment of the present invention.
Detailed Description
The following describes the embodiments of the present invention further with reference to the drawings. The description of these embodiments is provided to assist understanding of the present invention, but is not intended to limit the present invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
It should be noted that, in the description of the present invention, the positional or positional relation indicated by the terms such as "upper", "lower", "left", "right", "front", "rear", etc. are merely for convenience of describing the present invention based on the description of the structure of the present invention shown in the drawings, and are not intended to indicate or imply that the apparatus or element to be referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
The terms "first" and "second" in this technical solution are merely references to the same or similar structures, or corresponding structures that perform similar functions, and are not an arrangement of the importance of these structures, nor are they ordered, or are they of a comparative size, or other meaning.
In addition, unless explicitly stated and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., the connection may be a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two structures. It will be apparent to those skilled in the art that the specific meaning of the terms described above in this application may be understood in the light of the general inventive concept in connection with the present application.
Example 1
An airport runway foreign object detection method deployed on a computing device or server, as shown in fig. 1, comprising the steps of:
s1: an airport runway image is acquired.
In this embodiment, the image of the airport runway is obtained after the image acquisition device photographs the airport runway. In practical application, considering that the length and width of the runway are generally large, the image acquisition device is difficult to cover the runway at one time, so the image acquisition device shoots and obtains runway images in the following modes:
firstly, dividing an airport runway into a plurality of preset positions along the length and/or width direction of the airport runway;
and then the image acquisition equipment sequentially rounds the plurality of preset positions at preset scanning time, so that airport runway images about each preset position can be sequentially obtained.
The image acquisition equipment is configured as a camera, the camera can be deployed on a tower erected at the runway drainage ditch, the tower is usually provided with a plurality of cameras along the length direction of the airport runway, each camera is installed on each tower, and each camera is responsible for the image acquisition work of the airport runway with one section of length. The cameras are converged on a computing device or a server deployed with the method through a network. In this embodiment, taking 250m of tower from runway center line as an example, in accordance with civil aviation regulations, the height of the camera from the ground at this distance should not exceed 15, preferably 15m, in order to ensure flight safety. The camera selects a lens with a focal length of at least 800mm to meet the requirement of long-distance coverage through calculation; to meet the coverage range variation requirement, the camera selects a lens with at least 30 times of optical zooming; selecting a cradle head type camera for meeting the requirement of lens picture stabilization during long-focus working; in order to meet the requirement of identifying tiny objects, a camera with at least 400 ten thousand pixels is selected, and the CMOS size is not less than 1/1.8; in order to solve the problems of insufficient night illumination caused by weak illumination of only side lamps (60 meters apart) at night and clear shooting when the distance is far, the laser light filling is selected to be used. In order to enable the shooting mode of the camera to adapt to various weather environments, the camera is set to be in an automatic photosensitive mode, a color mode is used when the illumination condition is good, and a laser mode is used when the illumination environment is poor (such as night, rainy days and the like), so that clear runway pictures can be shot in various weather environments. .
In addition, since the size of the foreign object is generally small, in order to ensure no false negative, it is necessary to ensure that the foreign object can be clearly captured by the camera, that is, that the foreign object has enough pixels (15 pixels are set here, which are minimum pixel requirements that can be stably recognized based on video analysis) in the airport runway image, so that the camera needs to use a long focal length. After using the tele, the corresponding camera field angle will be smaller, i.e. the runway area covered by the camera under a single scanning position will be smaller, and the corresponding scanning position will be more needed to cover the covered runway area needed by a single tower. Meanwhile, the flight interval of the airport (the time interval when two flights occur in the same area of the runway) needs to be considered, and the scanning identification of all scanning bits needs to be completed within the time interval of two flights, namely, the corresponding scanning bits need to be fewer. Therefore, in terms of factors of multi-directional interaction and restriction such as focal length selection, scanning position setting and time interval, the embodiment sets and uses 1000mm long focus (can ensure that targets with the size of 4cm have enough pixels, and the rest focal length can be used for manual amplified foreign matter verification after foreign matter detection), namely the embodiment selects and uses 1000mm long focus laser holder. The number of corresponding scanning bits is 10 (can ensure that all runway areas needing to be identified by a single tower can be completely covered), and the identification duration of each scanning bit is 10 seconds (can ensure that enough time is available for alarm self-verification on the basis of identification). With this arrangement, a single full runway area scan takes about 100 seconds and a flight interval of at least about 120 seconds, i.e., the system ensures that airport runway debris is captured at the capture time.
It can be appreciated that in the process of acquiring the airport runway image by the image acquisition device, the image stabilization algorithm is applied to solve the problem of image shake.
S2: visual features are extracted from the airport runway images by a feature extractor.
S3: and processing visual features extracted from the airport runway images through a pre-trained foreign matter detection model to obtain an abnormal probability value of each position in the airport runway images.
Airport runway alien detection is different from the problems and tasks in the conventional mode, and is aimed at a few, unpredictable or uncertain rare events, has unique complexity and is a typical context abnormality triggered by unknown objects in space data.
The method based on representation mostly adopts a deep convolutional neural network to extract the image characteristics of normal samples, and the corresponding distribution is characterized by a non-parameter distribution estimation method, and the anomaly score is calculated by measuring the distance between the characteristics of the test image and the estimated distribution. Current methods do not efficiently map image features onto a base distribution that is easy to handle and ignore the relationship between local and global features, which are important to identify anomalies. The object types to be identified in the foreign matter detection are various, large to vehicles, small to screws and various scraps and parts, so that the conventional model training and identification mode for various foreign matters by stacking a large number of samples is utilized on the basis of a deep learning mode, the types of the foreign matters can be infinitely increased, and the mode obviously cannot meet the actual demands in infinite updating iteration of sample models.
Thus, in an embodiment, the FastFlow algorithm implemented in two-dimensional normalized flow is used and used as a probability distribution estimator, and can be used as a plug-in module with any depth feature extractor such as ResNet and visual converter for unsupervised anomaly detection and localization.
In the training stage, visual feature extraction is carried out on the airport runway sample image without foreign matters through a feature extractor; and training the FastFlow algorithm realized by the two-dimensional normalized flow by using the visual features extracted from the airport runway sample image, so that the FastFlow algorithm realized by the two-dimensional normalized flow can convert the input visual features into processable distribution, namely, the original distribution is converted into standard normal distribution in a 2D mode, and the abnormal probability value of each position (each position on the two-dimensional feature) in the airport runway sample image is obtained in an reasoning stage, thereby obtaining the required foreign object detection model. In the application stage, firstly, the visual features in the input airport runway images are extracted through a feature extractor, and then the visual features are input into a trained foreign matter detection model to estimate the abnormal probability value of each position in the airport runway images.
S4: and identifying the foreign matters in the airport runway images through the abnormal probability values to obtain a foreign matter detection result.
That is, for each position in the airport runway image, it is determined whether or not the abnormal probability value is higher than the sensitivity threshold, and if so, it is determined as a foreign object. By applying the method provided by the embodiment, in the actual detection of the foreign matters on the airport runway, the detection rate of the foreign matters on the airport runway exceeds 99%.
In step S1, the image acquisition device and the airport runway are also usually calibrated by calibration techniques in order to determine their position on the airport runway after the detection of the foreign bodies. In actual application, the server or the computing equipment deployed with the method can display the foreign matter detection result on a visual interface (comprising a display or a mobile phone interface) and give an alarm, and meanwhile, the position of the foreign matter is marked on the visual interface, so that airport management and control personnel can quickly locate the position of the foreign matter, further, the personnel is arranged to carry out emergency treatment, and meanwhile, if the management and control personnel find that the foreign matter is false alarm, alarm elimination operation can be carried out.
Example two
As shown in fig. 2, before the step of identifying the foreign matter in the airport runway image by the abnormal probability value, the steps of:
s31: whether the airfield runway image contains an aircraft is detected through a target detection algorithm, if yes, the process goes to S32, otherwise, the process goes to S33.
S32: and adjusting the sensitivity threshold by a preset amplitude reduction, thereby reducing the false alarm rate.
Partial false detection may occur due to changes in illumination, changes in various background text on the runway of the airport, and the influence of settings of side lights on the runway, etc. To solve such problems, the present embodiment adds an aircraft identification algorithm, i.e. when an aircraft is detected, the sensitivity threshold is reduced to avoid a large number of false alarms caused by illumination due to illumination of the aircraft at night and the aircraft itself also belongs to a foreign object (not a conventionally supposed target on the runway).
S33: and judging whether the acquisition time of the airfield runway image is within a preset period after detecting that the airfield runway image contains the aircraft, and if so, proceeding to S34.
S34: and adjusting the sensitivity threshold by preset amplification, thereby reducing the rate of missing report.
It will be appreciated that since the friction between the aircraft and the ground as well as the vibration of the aircraft itself during high speed movement will cause the aircraft itself to drop, thus forming a foreign object on the airport runway, the sensitivity threshold will need to be increased for a predetermined period of time after the aircraft is detected to be away (disappeared from the airport runway image) to ensure that there is no false alarm. After that (i.e., when the determination result of S33 is no), the sensitivity threshold is recovered, and it is not necessary to always maintain a higher sensitivity threshold.
Example III
As shown in fig. 3, the foreign matter detection result is obtained by further comprising the steps of:
and S5, tracking the detected foreign matters through a track tracking algorithm, so as to obtain the motion state of the foreign matters, and filtering false alarms based on the principle of the motionless foreign matters.
In the embodiment, the track tracking algorithm is utilized to track each detected foreign object (including suspected foreign objects), and the motion state of the foreign object is identified by utilizing the characteristic that the foreign object cannot move, so that false alarm targets corresponding to the detected target motion caused by unstable detection are filtered. Therefore, false alarms can be reduced as much as possible under the condition of high detection rate, and in the actual runway foreign matter detection, the false alarm rate of the foreign matter detection on the runway is lower than 5%.
Example IV
An airport runway foreign matter detection system, as shown in FIG. 4, comprises
The image acquisition module is used for receiving an airport runway image to be detected;
the feature extraction module is used for extracting visual features from the airport runway images;
the model calling module is used for calling a pre-trained foreign matter detection model to process the airport runway image and obtain an abnormal probability value of each position in the airport runway image;
and the foreign matter identification module is used for identifying the foreign matters in the airport runway images through the abnormal probability value to obtain a foreign matter detection result.
Example five
An electronic device, as shown in fig. 5, includes a memory storing executable program code and a processor coupled to the memory; wherein the processor invokes executable program code stored in the memory to perform the method steps disclosed in the above embodiments.
Example six
A computer storage medium having a computer program stored therein, which when executed by a processor performs the method steps disclosed in the above embodiments.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, and yet fall within the scope of the invention.
Claims (10)
1. An airport runway foreign matter detection method is characterized in that: the method comprises the following steps:
acquiring an airport runway image;
extracting visual features from the airport runway images by a feature extractor;
processing visual features extracted from the airport runway images through a pre-trained foreign matter detection model to obtain an abnormal probability value of each position in the airport runway images;
and identifying the foreign matters in the airport runway images through the abnormal probability value to obtain a foreign matter detection result.
2. The method according to claim 1, characterized in that: the pre-trained foreign matter detection model is obtained through the following steps:
visual feature extraction is carried out on the airport runway sample image without foreign matters through a feature extractor;
training a two-dimensional normalized flow implemented FastFlow algorithm using visual features extracted from the airport runway sample image, such that the two-dimensional normalized flow implemented FastFlow algorithm society converts the input visual features into processable distributions, and obtains an abnormal probability value for each position in the airport runway sample image in an inference stage, thereby obtaining the foreign object detection model.
3. The method according to claim 1, characterized in that: the step of obtaining the runway image comprises the following steps:
firstly dividing an airport runway into a plurality of preset positions; and sequentially carrying out round-robin on a plurality of preset positions at preset scanning time by using an image acquisition device, so as to obtain airport runway images about each preset position.
4. A method according to claim 3, characterized in that: the image acquisition equipment is a long-focus laser holder.
5. A method according to claim 3, characterized in that: and calibrating the image acquisition equipment and the airport runway through a calibration technology so as to determine the position of the foreign matters on the airport runway after the foreign matters are detected.
6. The method according to claim 1, characterized in that: the step of identifying a foreign object in the airport runway image by the abnormal probability value includes:
and judging whether the abnormal probability value of each position in the airport runway image is higher than a sensitivity threshold value, and judging that the position is foreign matter.
7. The method according to claim 6, wherein: when the airport runway image is detected to contain aircrafts, adjusting the sensitivity threshold value by a preset amplitude reduction, so that the false alarm rate is reduced; wherein whether the airport runway image contains an aircraft is detected through an object detection algorithm.
8. The method according to claim 7, wherein: and under the condition that the airport runway image is detected to contain the aircraft within a preset period of time and the airport runway image does not contain the aircraft, adjusting the sensitivity threshold value by preset amplification, so that the report missing rate is reduced.
9. The method according to claim 1, characterized in that: the method further comprises the following steps after the foreign matter detection result is obtained:
tracking the detected foreign matters through a track tracking algorithm, so as to obtain the motion state of the foreign matters, and filtering false alarms based on the principle of the foreign matters not moving.
10. An airport runway foreign matter detecting system which is characterized in that: comprising
The image acquisition module is used for receiving an airport runway image to be detected;
the feature extraction module is used for extracting visual features from the airport runway images;
the model calling module is used for calling a pre-trained foreign matter detection model to process the airport runway image and obtain an abnormal probability value of each position in the airport runway image;
and the foreign matter identification module is used for identifying the foreign matters in the airport runway images through the abnormal probability value to obtain a foreign matter detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310329815.3A CN116434142A (en) | 2023-03-30 | 2023-03-30 | Airport runway foreign matter detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310329815.3A CN116434142A (en) | 2023-03-30 | 2023-03-30 | Airport runway foreign matter detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116434142A true CN116434142A (en) | 2023-07-14 |
Family
ID=87086595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310329815.3A Pending CN116434142A (en) | 2023-03-30 | 2023-03-30 | Airport runway foreign matter detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116434142A (en) |
-
2023
- 2023-03-30 CN CN202310329815.3A patent/CN116434142A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103733234B (en) | A surveillance system and a method for detecting a foreign object, debris, or damage in an airfield | |
| CN107728136A (en) | A kind of airfield runway monitoring foreign bodies and removing guiding system and method | |
| US9483952B2 (en) | Runway surveillance system and method | |
| CN104971458B (en) | Based on the more fire source recognition methods for automatically tracking positioning jet stream extinguishing device | |
| CN109164443A (en) | Rail track foreign matter detecting method and system based on radar and image analysis | |
| CN104408932A (en) | Drunk driving vehicle detection system based on video monitoring | |
| CN104729483B (en) | Foreign bodies on transmission lines species detection platform | |
| CN105303844B (en) | Night expressway fog automatic detection device and its detection method based on laser | |
| CN104713526A (en) | Method for detecting types of foreign matters on power transmission line | |
| CN104182992B (en) | Method for detecting small targets on the sea on the basis of panoramic vision | |
| CN111898444A (en) | Aircraft landing gear state determination method based on image recognition | |
| CN116453278A (en) | Intrusion monitoring method combining deep learning intelligent detection and optical fiber vibration sensing | |
| CN110544271B (en) | Parabolic motion detection method and related device | |
| CN116434142A (en) | Airport runway foreign matter detection method and system | |
| CN207502720U (en) | A kind of airfield runway monitoring foreign bodies and removing guiding system | |
| CN114119713A (en) | Forest land baldness detection method based on artificial intelligence and unmanned aerial vehicle remote sensing | |
| Jia et al. | Rapid Identification of Power Supply Line Galloping in Night Vision Scene Based on YOLOv5 Model | |
| CN115100544A (en) | Power transmission line satellite-ground cooperative external damage monitoring and early warning method, device, equipment and medium | |
| AU2008293060B2 (en) | Runway surveillance system and method | |
| CN117710809A (en) | Intelligent detection method and system for building outer wall falling | |
| CN116259030A (en) | Grab ship unloader collision early warning system based on target detection and image processing | |
| CN111768414A (en) | Photoelectric rapid aiming method and device for laser countermeasure system | |
| Parida et al. | ROAD DETECTION REGION GROWING BY MARKOV RANDOM FIELD |
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 |