Low-slow small target locking identification method based on second-class superlattice infrared detector

Abstract

The invention relates to the technical field of target locking identification, in particular to a low-slow small target locking identification method based on a second-class superlattice infrared detector. According to the method, after a suspected target in an airspace is obtained, an infrared detector is used for detection, and an infrared frame image sequence of the suspected target in each detection time period of the detection process is obtained; the radiation change frequency of the suspected target in each detection time period is analyzed and acquired; and further analyzing the radiation change frequency and the position change of the suspected target to obtain a bird confidence parameter, and further judging whether the suspected target is a low-slow small target or not, and locking and observing. According to the method, infrared detection is carried out after the suspected target is preliminarily recognized, then the possibility that the suspected target is a bird is evaluated based on the infrared radiation and position change characteristics of the suspected target in the infrared detection process, finally whether the suspected target is a low, slow and small target or not is accurately judged, the misjudgment rate is reduced, and the locking recognition effect on the low, slow and small target is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of target locking and identification, and in particular to a method for locking and identifying a low-speed, slow, and small target based on a second-class superlattice infrared detector. Background Art

[0002] A type-II superlattice infrared detector is an infrared detector based on type-II superlattice materials with high-sensitivity infrared radiation detection capabilities; low, slow, and small targets usually refer to low-altitude, low-speed, weak-signal, and small-sized artificial flying targets, such as drones and aircraft. In restricted areas, low, slow, and small artificial flying targets will pose a threat to military air defense and national security, so accurate locking and identification of them is crucial.

[0003] Because small, slow, and low-flying targets can generate thermal radiation, radar is typically used for initial long-range detection and early warning in some restricted airspace areas, followed by identification and lock-on with a Type II superlattice infrared detector. However, flying targets such as birds may also appear in restricted airspace areas. Birds also generate thermal radiation, and during flight, they may be mistakenly locked onto and identified by the Type II superlattice infrared detector, thus reducing the effectiveness of locking and identifying small, slow, and low-flying targets. Summary of the Invention

[0004] In order to solve the technical problem of poor locking and identifying effects on low-speed, slow, and small targets, the present invention aims to provide a method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector. The technical solution adopted is as follows:

[0005] A low-speed, slow, and small target locking and identification method based on a second-type superlattice infrared detector comprises:

[0006] After obtaining the suspected target in the airspace, the suspected target is detected using an infrared detector. The position of the infrared detector remains unchanged during the detection process. In each detection period of the detection process, a sequence of infrared frames of the suspected target is obtained. The duration of all detection periods is the same.

[0007] According to the infrared radiation change of the suspected target in each infrared frame sequence, the radiation change frequency of the suspected target in each detection period is obtained; according to the difference in the radiation change frequency of the suspected target between all adjacent detection periods and the position difference between the corresponding infrared frame sequences in all adjacent detection periods, the bird confidence parameter is obtained;

[0008] According to the bird confidence parameter, it is determined whether the suspected target is a low, slow and small target and locked for observation.

[0009] Furthermore, the method for obtaining the radiation change frequency includes:

[0010] In each frame of each infrared frame sequence, the radiation intensity of the suspected target is obtained according to the infrared radiation values of all pixels in the local range corresponding to the suspected target;

[0011] According to the change of the radiation intensity of the suspected target in each infrared frame sequence, the radiation change period value of the suspected target in the corresponding detection period is obtained, and the inverse of the radiation change period value is used as the radiation change frequency in the corresponding detection period.

[0012] Furthermore, the method for obtaining the radiation variation period value includes:

[0013] Different cycle values are set in each detection period, and any cycle value is used as the target cycle value. The target cycle value is used to divide the corresponding infrared frame sequence into different subsequences of equal length; according to the difference in the change of the radiation intensity of the suspected target between different subsequences, the irregular change parameter under the target cycle value is obtained; among all the cycle values, the cycle value corresponding to the smallest irregular change parameter is used as the radiation change cycle value.

[0014] Furthermore, the method for setting the period value includes:

[0015] In each detection period, half of the total number of frames in the infrared frame sequence is used as the right endpoint of the period value interval, the constant 2 is used as the left endpoint of the period value interval, and all integers in the period value interval are used as period values.

[0016] Furthermore, the method for obtaining the irregularly changing parameter includes:

[0017] In each subsequence, the sum of the radiation intensities of the suspected targets in all frame images is used as a radiation reference value; and the variance of the radiation reference values of all subsequences is used as an irregular change parameter.

[0018] Furthermore, the method for obtaining the bird confidence parameter includes:

[0019] Obtaining the wingbeat parameters of the suspected target in each detection period according to the radiation change frequency of the suspected target in each detection period and the change amount of the radiation change frequency in the previous detection period;

[0020] The position coordinates of the suspected target in each detection period are obtained based on the geometric center coordinates of the suspected target in all frames of the infrared frame sequence; the liftoff parameters of the suspected target in each detection period are obtained based on the vertical coordinate change of the position coordinates of the suspected target in each detection period relative to the position coordinates of the previous detection period;

[0021] According to the same sign of the values of the wing-flapping parameter and the lift-off parameter, the bird confidence sub-parameter of the suspected target in each detection period is obtained; and the bird confidence parameter of the suspected target in all detection periods is integrated to obtain the bird confidence parameter of the suspected target.

[0022] Furthermore, the method for obtaining the bird confidence sub-parameter includes:

[0023] The flapping parameter is used as the numerator, the lift-off parameter is used as the denominator, and the fraction ratio is positively correlated and mapped as the bird confidence sub-parameter; when the lift-off parameter is equal to 0, the bird confidence sub-parameter is set to 0.

[0024] Furthermore, the screening method for low, slow and small targets includes:

[0025] The bird confidence parameter is normalized. When the normalized value is greater than a preset threshold, the corresponding suspected target is determined to be a bird, otherwise it is a low, slow and small target.

[0026] Furthermore, the method for obtaining the suspected target includes:

[0027] A radar echo signal in the airspace is acquired by using a radar, and a Doppler frequency shift of the radar echo signal is acquired. When the Doppler frequency shift is greater than a preset frequency shift threshold, it is determined that a suspected target appears in the echo direction of the radar echo signal.

[0028] Furthermore, the detection period lasts for 1 second.

[0029] The present invention has the following beneficial effects:

[0030] After obtaining a suspected target in the airspace, the present invention uses an infrared detector to detect the suspected target. Then, in each detection period of the detection process, an infrared frame image sequence of the suspected target is obtained, providing a data analysis basis for the subsequent analysis of the bird confidence parameter of the suspected target. Then, based on the infrared radiation change of the suspected target in each infrared frame image sequence, the radiation change frequency of the suspected target in each detection period is obtained to help evaluate whether the suspected target has the wing-flapping characteristics of a bird. Further, based on the difference in the radiation change frequency of the suspected target between all adjacent detection periods and the position difference between the corresponding infrared frame image sequences in all adjacent detection periods, the suspected target is analyzed to determine whether it has the characteristic of accelerated wing-flapping frequency when rising, thereby evaluating and obtaining the bird confidence parameter. Finally, based on the bird confidence parameter, the suspected target is judged to be a low-speed, small target and locked for observation, reducing the possibility of misjudgment caused by interference from birds. The present invention preliminarily identifies the suspected target and then continuously monitors the suspected target using an infrared detector. Based on the infrared radiation change characteristics of the suspected target during the detection process, the possibility of the suspected target being a bird is assessed. Finally, the suspected target is accurately determined to be a low-speed, small target, reducing the misjudgment rate and improving the locking and identification effect of the low-speed, small target. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0032] Figure 1 A flow chart of a method for locking and identifying a low-speed, slow, and small target based on a second-class superlattice infrared detector provided by one embodiment of the present invention;

[0033] Figure 2 A flow chart of a method for acquiring radiation change frequency provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0034] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a method for locking and identifying a low, slow, and small target based on a second-class superlattice infrared detector proposed by the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0035] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0036] The following describes in detail a specific solution of a low-speed, slow, and small target locking and identification method based on a second-type superlattice infrared detector provided by the present invention with reference to the accompanying drawings.

[0037] See also Figure 1 , which shows a flow chart of a method for locking and identifying a low-speed, slow, small target based on a second-class superlattice infrared detector provided by one embodiment of the present invention, specifically including:

[0038] Step S1: After obtaining the suspected target in the airspace, the suspected target is detected using an infrared detector. The position of the infrared detector remains unchanged during the detection process. In each detection period of the detection process, an infrared frame image sequence of the suspected target is obtained. The duration of all detection periods is the same.

[0039] It should be noted that the implementation scenario targeted by the present invention is to first capture the suspected target in the airspace, then use the second-type superlattice infrared detector to track and detect the suspected target, and further analyze and evaluate the possibility that the suspected target is a bird, so as to accurately identify low, slow and small targets.

[0040] Considering that the radar can detect targets at a relatively long distance and obtain the approximate direction and speed of the target, it helps to preliminarily identify potential threats and guide the working direction of other sensors; after obtaining the approximate direction of the target, the second-type superlattice infrared detector can further capture the target's infrared radiation and track the target, and then accurately lock and identify low, slow and small targets.

[0041] Based on this, in one embodiment of the present invention, a detection radar, such as a millimeter-wave radar, is first installed at an airspace monitoring point. The radar's rotation speed is set to once per second, that is, the airspace monitoring frequency of the detection radar is also once per second. During the radar's rotation, the radar transmitter emits electromagnetic waves in every direction within a 360-degree range of the monitored airspace every second. When a suspected target is detected, an echo is reflected, thereby obtaining the suspected target in the airspace.

[0042] Preferably, in one embodiment of the present invention, considering that low, slow, and small targets are usually mobile, if the airspace object moves relative to the radar, that is, when the airspace object is moving, the frequency of the electromagnetic wave signal reflected by the airspace object received by the radar, that is, the radar echo signal, will change. This frequency change is called Doppler shift. Therefore, based on the Doppler shift, it can be determined whether there is a suspected low, slow, and small target in the airspace. The method for obtaining the suspected target includes:

[0043] The radar echo signal in the airspace is acquired by using the radar, and the Doppler frequency shift of the radar echo signal is obtained. When the Doppler frequency shift is greater than the preset frequency shift threshold, it is determined that a suspected target appears in the echo direction of the radar echo signal.

[0044] As an example, the radar echo signal is subjected to certain preprocessing, such as amplification and filtering, to further obtain its Doppler frequency shift; the preset frequency shift threshold is set to 50Hz, which can also be customized by the implementer; when the Doppler frequency shift is greater than 50Hz, it is considered that the airspace object in the echo direction of the radar echo signal has motion characteristics and is a low, slow and small suspected target; the implementer can also further determine the direction of the suspected target based on the radar echo signal for subsequent tracking and detection by the second-type superlattice infrared detector.

[0045] It should be noted that the preprocessing of the radar echo signal, the acquisition of the Doppler frequency shift, and the determination of the direction of the suspected target based on the radar echo signal are all existing technologies well known to those skilled in the art and will not be described in detail.

[0046] In one embodiment of the present invention, a second-type superlattice infrared detector is synchronously set at an airspace monitoring point. When a suspected target in the airspace is captured, the second-type superlattice infrared detector starts to work; then, the position of the second-type superlattice infrared detector is adjusted according to the orientation of the suspected target so that the center of the infrared image coincides with the geometric center of the suspected target; the second-type superlattice infrared detector is used to lock and continuously detect the suspected target, and the detection process is not less than 5 seconds, and the position of the infrared detector remains unchanged during the detection process; wherein, the suspected target may move during the detection process, and the image center of the subsequent infrared frame image may no longer coincide with the geometric center of the suspected target;

[0047] The detection process is then divided into several detection periods of equal length. For example, the duration of the detection period is set to 1 second, and all detection periods in the detection process are obtained. The detection frequency of the second-type superlattice infrared detector is further set to 30 Hz, that is, 30 frames of infrared frame images are obtained in each detection period, and then the frames are sorted in the acquisition order to construct an infrared frame sequence.

[0048] It should be noted that adjusting the position of the second-type superlattice infrared detector according to the orientation of the suspected target is an existing technology well known to those skilled in the art and will not be described in detail; in other embodiments, the implementer can also define the duration and detection frequency of the detection period.

[0049] It should be noted that the deployment and application of detection radars and Class II superlattice infrared detectors in airspace monitoring points are already well-known technologies, and implementers can adjust the deployment plan according to actual conditions. There may also be multiple suspected targets in the airspace at the same time. In this case, multiple Class II superlattice infrared detectors can be set in combination with the detection viewing angle range of the Class II superlattice infrared detector to ensure that the entire airspace can be covered, and then multiple suspected targets can be tracked at the same time. Among them, a Class II superlattice infrared detector can also observe multiple suspected targets within the viewing angle range at the same time.

[0050] It should be noted that the method for locking and identifying whether each suspected target is a low, slow, and small target is the same. Here, only any suspected target is used as an example for analysis and description.

[0051] Step S2: According to the infrared radiation change of the suspected target in each infrared frame sequence, the radiation change frequency of the suspected target in each detection period is obtained; according to the difference in the radiation change frequency of the suspected target between all adjacent detection periods and the position difference between the corresponding infrared frame sequences in all adjacent detection periods, the bird confidence parameter is obtained.

[0052] Considering that man-made flying targets in the airspace usually exhibit infrared radiation characteristics, monitoring the infrared radiation of suspected targets can help identify low, slow, and small targets. Considering that birds also exhibit infrared radiation characteristics when flying in the airspace, but birds need to flap their wings to maintain flight, and the infrared radiation of birds will show a certain regular change with the flapping of their wings, while the infrared radiation of man-made flying targets is relatively stable.

[0053] Therefore, the embodiment of the present invention will first analyze the radiation change frequency of the suspected target in each detection period based on the infrared radiation change of the suspected target in each infrared frame sequence, and then facilitate the subsequent evaluation of the bird confidence parameter based on the distinguishing characteristics of the infrared radiation change of birds and the infrared radiation change of small targets that are relatively slow and slow.

[0054] Preferably, in one embodiment of the present invention, the method for obtaining the radiation change frequency includes:

[0055] See also Figure 2 , which shows a flow chart of a method for obtaining radiation change frequency provided by an embodiment of the present invention, specifically comprising:

[0056] Step S201 : in each frame of each infrared frame sequence, the radiation intensity of the suspected target is obtained according to the infrared radiation values of all pixels in the local range corresponding to the suspected target.

[0057] To facilitate subsequent analysis of changes in the infrared radiation intensity of the suspected target, the embodiment of the present invention first obtains the radiation intensity of the suspected target in each frame of the infrared frame sequence corresponding to each detection period.

[0058] In one embodiment of the present invention, it is first necessary to obtain the suspected target corresponding area in each infrared frame image sequence; the implementer can obtain the suspected target corresponding pixel points in each frame image based on threshold judgment; the threshold is set to the mean or mode of all pixel points in the frame image, and when the pixel value is greater than the threshold, the corresponding pixel point is determined to be the suspected target corresponding pixel point, and the connected domain of all suspected target corresponding pixel points is the suspected target corresponding area.

[0059] In other embodiments, the implementer may also manually label the suspected targets in the frame images, and then construct a training set to train a labeling model for the suspected targets, and then input the frame images to obtain the suspected targets in the frame images; other methods such as threshold segmentation or first obtaining the suspected targets in the first frame image, and then using the optical flow method to obtain the suspected targets in all frame images may also be used; these are all existing technologies and will not be repeated here.

[0060] After obtaining the suspected target in each frame image, the radiation intensity of the suspected target can be further obtained.

[0061] As an example, the pixel values, that is, the average infrared radiation values, of all pixels in the corresponding area of each frame image of the suspected target are used as the radiation intensity of the suspected target in the frame image; in other examples, the implementer may also expand two unit pixels in the outermost layer of the area corresponding to the suspected target, that is, expand twice from the outermost layer to obtain the target expansion area, and further use the average pixel values of all pixels in the target expansion area as the radiation intensity of the suspected target in the corresponding frame image.

[0062] Step S202: According to the change of the radiation intensity of the suspected target in each infrared frame sequence, the radiation change period value of the suspected target in the corresponding detection period is obtained, and the inverse of the radiation change period value is used as the radiation change frequency in the corresponding detection period.

[0063] In order to facilitate the subsequent evaluation of whether it has the characteristics of bird wing flapping and determine the bird confidence coefficient, the embodiment of the present invention will analyze the infrared radiation change pattern of the suspected target in each detection period, and obtain the radiation change period value of the suspected target in the corresponding detection period. The larger the radiation change period value, the lower the radiation change frequency, and then the possibility of bird can be evaluated in combination with the frequency characteristics of bird wing flapping.

[0064] In a preferred embodiment of the present invention, if the radiation intensity of a suspected target varies periodically, the radiation intensity in different periods will be similar. The period is divided into sub-segments based on a preset period value, and the irregular variation of the radiation intensity of the suspected target in different period sub-segments is evaluated, which can help evaluate the confidence level of the corresponding preset period. Then, by iterating different preset period values, the radiation variation period value is analyzed and evaluated. Based on this, the method for obtaining the radiation variation period value includes:

[0065] Different cycle values are set in each detection period, and any cycle value is taken as the target cycle value. The target cycle value is used to divide the corresponding infrared frame sequence into different subsequences of equal length. According to the difference in the change of radiation intensity of suspected targets between different subsequences, the irregular change parameter under the target cycle value is obtained. Among all the cycle values, the cycle value corresponding to the smallest irregular change parameter is taken as the radiation change cycle value.

[0066] In a preferred embodiment of the present invention, considering that the length of the subsequence should be at least greater than 1 and not too long to avoid invalid analysis, the method for setting the period value includes:

[0067] In each detection period, half of the total number of frames in the infrared frame sequence is used as the right endpoint of the period value interval, the constant 2 is used as the left endpoint of the period value interval, and all integers in the period value interval are used as period values; that is, the period value interval is from 2 to half of the total number of frames, and is a double-closed interval; implementers can also set the period value set for iterative analysis.

[0068] In a preferred embodiment of the present invention, considering that there are multiple frames of images in each subsequence, in order to facilitate the analysis of the difference in radiation intensity of suspected targets in different subsequences, the radiation intensity of the suspected targets in all frames in each subsequence is first integrated to determine a radiation reference value. Furthermore, considering that the variance can measure the discreteness of the radiation reference values between different subsequences, it helps to evaluate the irregular changes in the radiation intensity of the suspected targets during the detection period. Based on this, the method for obtaining the irregular change parameter includes:

[0069] In each subsequence, the sum of the radiation intensities of the suspected targets in all frames is taken as the radiation reference value; the variance of the radiation reference values of all subsequences is taken as the irregular change parameter.

[0070] In other embodiments, the implementer may also construct a corresponding radiation intensity sequence by sorting the radiation intensity of the suspected target in each frame image of each subsequence according to the order of the frame images; then compare the differences between the radiation intensity sequences of each subsequence, such as the DTW distance, and then use the sum of all differences obtained by the pairwise comparison as the irregular change parameter; this is already existing technology and will not be repeated here.

[0071] Considering that when birds fly in the airspace, the frequency of their wings flapping will gradually increase in order to maintain the buoyancy of the ascent, the frequency of radiation changes of birds corresponding to suspected targets in adjacent detection periods will gradually increase during the detection process; when descending, birds usually glide to save energy, and the frequency of wing flapping will decrease or stop flapping, so the frequency of radiation changes of birds corresponding to suspected targets in adjacent detection periods should gradually decrease; and when ascending or descending, the position of the suspected target in the frame image will change;

[0072] Based on this, an embodiment of the present invention obtains a bird confidence parameter based on the difference in the radiation change frequency of the suspected target between all adjacent detection time periods, and the position difference between the corresponding infrared frame image sequences in all adjacent detection time periods; the bird confidence parameter reflects the possibility that the suspected target has bird characteristics, preparing for the subsequent judgment that the suspected target is a low, slow and small target.

[0073] Preferably, in one embodiment of the present invention, considering that the radiation change rate of the suspected target during the detection period shows an increasing trend relative to the previous detection period, it indicates that the suspected target may have a characteristic trend of increasing bird wingbeat frequency to float upward; and the change in the vertical coordinate of the geometric center of the suspected target in the infrared frame image can help assess whether the suspected target is floating upward; if the wingbeat frequency increases and the vertical coordinate increases, the possibility that the suspected target is a bird is greater; based on this, the method for obtaining the bird confidence parameter includes:

[0074] Obtain the wingbeat parameters of the suspected target in each detection period based on the radiation change frequency of the suspected target in each detection period and the change amount of the radiation change frequency in the previous detection period;

[0075] The position coordinates of the suspected target in each detection period are obtained based on the geometric center coordinates of the suspected target in all frames of the infrared frame sequence; the liftoff parameters of the suspected target in each detection period are obtained based on the vertical coordinate change of the position coordinates of the suspected target in each detection period relative to the position coordinates of the previous detection period;

[0076] According to the same sign of the wing flapping parameters and the lift-off parameters, the bird confidence sub-parameters of the suspected target in each detection period are obtained; the bird confidence parameters of the suspected target in all detection periods are integrated to obtain the bird confidence parameters of the suspected target.

[0077] In a preferred embodiment of the present invention, the method for obtaining the bird confidence sub-parameter includes:

[0078] The flapping parameter is used as the numerator, the lift-off parameter is used as the denominator, and the fraction ratio is positively correlated and mapped as the bird confidence sub-parameter; when the lift-off parameter is equal to 0, the bird confidence sub-parameter is set to 0.

[0079] As an example, first pass The change in the radiation change frequency of the suspected target is calculated in this way, and this value is used as the wingbeat parameter of the suspected target in the i+1th detection period; where i is the serial number of the detection period, w i is the radiation change frequency of the suspected target in the i-th detection period; w i+1 is the radiation change frequency of the suspected target in the i+1th detection period; similarly, the lift-off parameters of the suspected target in the i+1th detection period can be obtained; then the ratio of the wingbeat parameter to the lift-off parameter of the suspected target in each detection period is used as the x in the power function exp(x) with the natural constant e as the base, and the value range is adjusted with positive correlation; when the numerator and denominator have different signs, the smaller the mapping result, the smaller the bird confidence sub-parameter; finally, the bird confidence sub-parameters of all detection periods are summed and linearly normalized to obtain the bird confidence parameter.

[0080] It should be noted that the wing-flapping parameters and the lift-off parameters may take negative values. When they take negative values, it means that the wing-flapping frequency of the suspected target decreases and the position drops.

[0081] In other examples, the implementer may also, in each detection period, when the wing flapping parameter and the lift-off parameter have the same sign, that is, when they are both positive or negative, record the bird confidence sub-parameter in that detection period as 1, otherwise it is 0; finally, accumulate the bird confidence sub-parameters in all detection periods during the detection process and normalize them to obtain the bird confidence parameter.

[0082] Step S3: Determine whether the suspected target is a low, slow, small target based on the bird confidence parameter and lock it for observation.

[0083] After obtaining the bird confidence parameters of the suspected target during the detection process, we can further determine whether it is a bird, which makes it easier to subsequently lock and identify whether a low, slow, and small target is a potential threat.

[0084] Preferably, in one embodiment of the present invention, the method for screening low, slow and small targets includes:

[0085] The bird confidence parameter is normalized. When the normalized value is greater than the preset threshold, the corresponding suspected target is determined to be a bird, otherwise it is a low, slow and small target. The preset threshold is set to 0.65, which can also be customized by the implementer.

[0086] In one embodiment of the present invention, when a suspected target is determined to be a low, slow, and small target, the low, slow, and small target is locked, and then it is continuously locked and observed based on a radar or a type-II superlattice infrared detector, and its flight trajectory in the airspace is recorded; then a large number of historical flight trajectories of low, slow, and small targets are obtained, and a spatial trajectory prediction model is constructed and trained, wherein the basic architecture of the model is an LSTM model; the spatial coordinate sequence corresponding to the current flight trajectory of the low, slow, and small target is input into the trained spatial model, and the predicted flight trajectory is output; the implementer can further obtain a large number of historical flight trajectories of low, slow, and small targets that have been manually marked with flight purposes or threat levels, train a threat assessment model, and thereby input the current flight trajectory of the low, slow, and small target into the trained threat assessment model, output its threat level, and then warn whether it is necessary to attack or disperse it.

[0087] It should be noted that the training and application of the above models are existing technologies and will not be described in detail.

[0088] In summary, after the present invention obtains the suspected target in the airspace, it uses an infrared detector to detect the suspected target, and the position of the infrared detector remains unchanged during the detection process; in each detection period of the detection process, an infrared frame image sequence of the suspected target is obtained, and the duration of all detection periods is the same; according to the infrared radiation change of the suspected target in each infrared frame image sequence, the radiation change frequency of the suspected target in each detection period is obtained; according to the difference in the radiation change frequency of the suspected target between all adjacent detection periods, and the position difference between the corresponding infrared frame image sequences in all adjacent detection periods, the bird confidence parameter is obtained; according to the bird confidence parameter, it is judged whether the suspected target is a low, slow, small target and locked for observation. After the present invention preliminarily identifies the suspected target, it uses an infrared detector to continuously detect the suspected target, and evaluates the possibility of the suspected target being a bird based on the infrared radiation change characteristics of the suspected target during the detection process, and finally accurately determines whether the suspected target is a low, slow, small target, reduces the misjudgment rate, and improves the locking and identification effect of the low, slow, small target.

[0089] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0090] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A low-speed, slow, and small target locking and identification method based on a second-class superlattice infrared detector is characterized by: The method comprises: After obtaining the suspected target in the airspace, the suspected target is detected using an infrared detector. The position of the infrared detector remains unchanged during the detection process. In each detection period of the detection process, a sequence of infrared frames of the suspected target is obtained. The duration of all detection periods is the same. According to the infrared radiation change of the suspected target in each infrared frame sequence, the radiation change frequency of the suspected target in each detection period is obtained; according to the difference in the radiation change frequency of the suspected target between all adjacent detection periods and the position difference between the corresponding infrared frame sequences in all adjacent detection periods, the bird confidence parameter is obtained; According to the bird confidence parameter, it is determined whether the suspected target is a low, slow and small target and locked for observation.

2. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 1 is characterized in that: The method for obtaining the radiation change frequency includes: In each frame of each infrared frame sequence, the radiation intensity of the suspected target is obtained according to the infrared radiation values of all pixels in the local range corresponding to the suspected target; According to the change of the radiation intensity of the suspected target in each infrared frame sequence, the radiation change period value of the suspected target in the corresponding detection period is obtained, and the inverse of the radiation change period value is used as the radiation change frequency in the corresponding detection period.

3. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 2 is characterized in that: The method for obtaining the radiation variation period value includes: Different cycle values are set in each detection period, and any cycle value is used as the target cycle value. The target cycle value is used to divide the corresponding infrared frame sequence into different subsequences of equal length; according to the difference in the change of the radiation intensity of the suspected target between different subsequences, the irregular change parameter under the target cycle value is obtained; among all the cycle values, the cycle value corresponding to the smallest irregular change parameter is used as the radiation change cycle value.

4. The method for locking and identifying a low-speed, slow, small target based on a second-class superlattice infrared detector according to claim 3 is characterized in that: The method for setting the period value includes: In each detection period, half of the total number of frames in the infrared frame sequence is used as the right endpoint of the period value interval, the constant 2 is used as the left endpoint of the period value interval, and all integers in the period value interval are used as period values.

5. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 3 is characterized in that: The method for obtaining the irregularly changing parameter includes: In each subsequence, the sum of the radiation intensities of the suspected targets in all frame images is used as a radiation reference value; and the variance of the radiation reference values of all subsequences is used as an irregular change parameter.

6. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 1 is characterized in that: The method for obtaining the bird confidence parameter includes: Obtaining the wingbeat parameters of the suspected target in each detection period according to the radiation change frequency of the suspected target in each detection period and the change amount of the radiation change frequency in the previous detection period; The position coordinates of the suspected target in each detection period are obtained based on the geometric center coordinates of the suspected target in all frames of the infrared frame sequence; the liftoff parameters of the suspected target in each detection period are obtained based on the vertical coordinate change of the position coordinates of the suspected target in each detection period relative to the position coordinates of the previous detection period; According to the same sign of the values of the wing-flapping parameter and the lift-off parameter, the bird confidence sub-parameter of the suspected target in each detection period is obtained; and the bird confidence parameter of the suspected target in all detection periods is integrated to obtain the bird confidence parameter of the suspected target.

7. The method for locking and identifying a low-speed, slow, small target based on a second-class superlattice infrared detector according to claim 6 is characterized in that: The method for obtaining the bird confidence sub-parameters includes: The flapping parameter is used as the numerator, the lift-off parameter is used as the denominator, and the fraction ratio is positively correlated and mapped as the bird confidence sub-parameter; when the lift-off parameter is equal to 0, the bird confidence sub-parameter is set to 0.

8. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 1 is characterized in that: The screening method for the low, slow and small targets includes: The bird confidence parameter is normalized. When the normalized value is greater than a preset threshold, the corresponding suspected target is determined to be a bird, otherwise it is a low, slow and small target.

9. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 1, characterized in that: The method for obtaining the suspected target includes: A radar echo signal in the airspace is acquired by using a radar, and a Doppler frequency shift of the radar echo signal is acquired. When the Doppler frequency shift is greater than a preset frequency shift threshold, it is determined that a suspected target appears in the echo direction of the radar echo signal.

10. The method for locking and identifying low-speed, slow, and small targets based on a second-class superlattice infrared detector according to claim 1, characterized in that: The duration of the detection period is 1 second.

Images