Multi-target thermal image bullet grabbing method and system

Abstract

The invention relates to the field of shooting training, and provides a multi-target thermal image bullet grabbing method and system.The method comprises the steps that multiple target targets are captured and tracked through thermal imaging equipment, and morphological parameters of the target targets and temperature data of all points are obtained; determining the target type of the target target according to the morphological parameters of the target target; according to the temperature data of each point of the target target, obtaining each abnormal point with abnormal temperature change on the target target; all abnormal points are removed and merged, and a bullet point area conforming to the shooting shape is obtained; according to the target type of the target target and the position of the bullet point area on the target target, the shooting score of the target target is obtained. The thermal imaging equipment can capture and track a plurality of target targets, capture the target targets appearing in the imaging range in time, screen out bullet point areas conforming to the shooting shape on the target targets, obtain the shooting scores of the target targets according to the positions of the bullet point areas on the target targets and the target types of the target targets, and improve the shooting quality of the target targets. And real-time target reading is realized.

Description

Technical Field

[0001] The invention relates to the field of shooting training, and further to a multi-target thermal imaging bullet capturing method and system. Background Art

[0002] In the existing shooting training of fixed targets, four points need to be burned around the target to mark the range of the target, so as to facilitate the scanning of the thermal imaging equipment and generate a background image in the system. However, in the existing shooting training, most of the targets are upside down targets and sideways targets, that is, the targets are moving, and it is difficult to read the shooting score on the moving targets in time. Summary of the invention

[0003] In view of the above technical problems, the purpose of the present invention is to provide a multi-target thermal imaging bullet capture method and system, in which the thermal imaging equipment can capture and track multiple target targets, timely capture the target targets that appear in the imaging range, and determine the target type of the target target according to the morphological data of the target target, and then screen out the bullet point area that meets the shooting shape on the target target, and obtain the shooting score of the target target in combination with the position of the bullet point area on the target target, and can read the target in real time for the upside-down target and the flipped target.

[0004] In order to achieve the above object, the present invention provides a multi-target thermal imaging capture method, comprising:

[0005] Capturing and tracking multiple targets by thermal imaging equipment to obtain morphological parameters of the target and temperature data of each point;

[0006] Determining the target type of the target according to the morphological parameters of the target;

[0007] According to the temperature data of each point of the target, each abnormal point with abnormal temperature change on the target is obtained;

[0008] All the abnormal points are removed and merged to obtain a bullet point area that conforms to the shooting shape;

[0009] The shooting score of the target target is obtained according to the target type of the target target and the position of the bullet point area on the target target.

[0010] In some embodiments, determining the target type of the target according to the morphological parameters of the target specifically includes:

[0011] Use artificial intelligence to train the morphological recognition of existing targets and put the morphological data of existing targets into the database;

[0012] Summarize and classify the morphological parameters of existing targets and obtain several target types of existing targets;

[0013] The morphological parameters of the target are obtained and compared with the morphological data of each type of target of the existing targets, and the target type of the target is determined.

[0014] In some embodiments, the morphology recognition training of existing targets by artificial intelligence specifically includes:

[0015] Set the existing target within the field of view of the thermal imaging device;

[0016] Adjust the placement position and placement angle of the existing target to obtain the video stream of the existing target;

[0017] Through artificial intelligence, feature extraction is performed on the video stream of the existing target to obtain feature data of the existing target, and then feature value stability calculation is performed on the feature data;

[0018] The feature data of existing targets are integrated into morphological data, and the morphological data is saved in the database.

[0019] In some embodiments, according to the temperature data of each point of the target, obtaining each abnormal point of the target where the temperature changes abnormally includes:

[0020] Obtaining the maximum temperature of the background image and filling the background image with the maximum temperature;

[0021] Acquire temperature data of each point of the target at a certain time point and a zero input response value and a zero state response value of the target, and calculate predicted change values ​​of each point of the target;

[0022] Acquire the temperature data of each point of the target at the next time point, and calculate the actual change value of each point of the target;

[0023] The predicted change value and the actual change value of each point of the target target are compared. If the actual change value of this point of the target target is much larger than the predicted change value of the target target, then this point is an abnormal point.

[0024] In some embodiments, the zero input response value of the target is ΔT=(T b +273)×e -αt , T b is the object temperature, α is the temperature change coefficient, t is the elapsed time, and e is the base of the natural exponent;

[0025] The zero-state response value of the target ΔT=(T c +273)×e -αt +d,T cis the ambient temperature, α is the temperature variation coefficient, t is the elapsed time, d is the temperature rise caused by the energy obtained from the infrared light in the sunlight, and e is the base of the natural index.

[0026] In some implementations, removing and merging all of the abnormal points to obtain a bullet point area that conforms to the shooting shape specifically includes:

[0027] Tracking and locating the abnormal point in time and space dimensions;

[0028] Removing the abnormal points that are isolatedly distributed in the spatial dimension;

[0029] Removing the abnormal points that appear briefly in the time dimension;

[0030] Merging a plurality of the abnormal points that are correlated in space and time into a target area;

[0031] The target area is subjected to a morphological analysis, and the target area that conforms to the shooting shape is the bullet point area.

[0032] In some implementations, the tracking and positioning of the abnormal point needs to be combined with the tracking of the target, and then the displacement law of the bullet point area position is calculated to achieve the correct positioning, prediction and tracking of the bullet point area.

[0033] In some embodiments, the thermal imaging device distinguishes the target from the background based on the difference in texture or temperature between the target and the external environment.

[0034] According to another aspect of the present application, a multi-target thermal imaging capture system is further provided, using any one of the multi-target thermal imaging capture methods of the above preferred embodiments, the multi-target thermal imaging capture system includes a thermal imaging device, an analysis module, a first calculation module, a screening module and a second calculation module, and the thermal imaging device can identify and track multiple targets;

[0035] The analysis module can analyze the morphological parameters of the target and obtain the target type of the target;

[0036] The first calculation module is capable of calculating the temperature change data of each point on the target and obtaining the abnormal points on the target;

[0037] The screening module can screen each abnormal point on the target and obtain the bullet point area;

[0038] The second calculation module is capable of calculating the shooting score of the bullet point area on the target.

[0039] In some embodiments, a training module is further included, wherein the training module is capable of performing morphology recognition training on existing targets and classifying the existing targets into several target types.

[0040] Compared with the prior art, the multi-target thermal imaging capture method and system provided by the present invention has at least one of the following beneficial effects:

[0041] 1. Thermal imaging equipment can capture and track multiple targets, timely capture the target targets that appear in the imaging range, and determine the target type of the target based on the morphological data of the target, and then select the bullet point area that meets the shooting shape on the target, and combine the position of the bullet point area on the target to obtain the shooting score of the target, and can read the target in real time for the upside-down target and the flip target.

[0042] 2. Determine the target type of the target through the morphological parameters of the target, and then determine the scoring criteria of the target.

[0043] 3. All abnormal points are removed and merged to obtain the bullet point area that conforms to the shooting shape, avoiding incorrect capture of the bullet point area and reading the wrong shooting score.

[0044] 4. The multi-target thermal imaging capture system can automatically capture and track multiple targets, dynamically screen and clean up interfering abnormal points, and read the shooting scores on the target more accurately and timely. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The preferred implementation modes will be described below in a clear and understandable manner with reference to the accompanying drawings to further illustrate the above-mentioned characteristics, technical features, advantages and implementation methods of the present invention.

[0046] Figure 1 It is a flow chart of a multi-target thermal imaging capture method;

[0047] Figure 2 It is a training graph for artificial intelligence morphological recognition;

[0048] Figure 3 It is a calculation flow chart of zero input response value and zero state response value;

[0049] Figure 4 is the screening graph of singular points. DETAILED DESCRIPTION

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings and other implementation methods can be obtained based on these drawings without creative work.

[0051] In order to simplify the drawings, only the parts related to the invention are schematically shown in each figure, and they do not represent the actual structure of the product. In addition, in order to simplify the drawings and facilitate understanding, in some figures, only one of the parts with the same structure or function is schematically drawn or marked. In this article, "one" not only means "only one", but also means "more than one".

[0052] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0053] In this document, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0054] In addition, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance. It should be noted that the above embodiments can be freely combined as needed. The above are only preferred implementations of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be regarded as the scope of protection of the present invention.

[0055] refer to Figure 1 The present invention provides a multi-target thermal imaging bullet capture method, which captures and tracks multiple target targets through thermal imaging equipment, timely captures the target targets that appear in the imaging range, and determines the target type of the target target according to the morphological data of the target target, and then selects the bullet point area that meets the shooting shape on the target target, and obtains the shooting score of the target target in combination with the position of the bullet point area on the target target, and can read the target in real time for the upside-down target and the flipped target.

[0056] A multi-target thermal imaging capture method specifically includes:

[0057] S101 captures and tracks multiple target targets through a thermal imaging device to obtain the morphological parameters of the target targets and the temperature data of each point. In the present embodiment, the thermal imaging device distinguishes the target targets from the background according to the texture or temperature difference between the target targets and the external environment. This relates to the principle of the thermal imaging device, which will not be further elaborated in this application. When the target targets appear within the field of view of the thermal imaging device, the thermal imaging device can automatically capture and track the target targets, and timely record the morphological parameters of the target targets and the temperature data of each point.

[0058] S102 determines the target type of the target according to the morphological parameters of the target.

[0059] In this embodiment, the target type of the target target is determined by the morphological parameters of the target target, and then the scoring standard of the target target is determined. Specifically, according to the morphological parameters of the target target, the target type of the target target is determined specifically including: S201 performing morphological recognition training on the existing target through artificial intelligence, and placing the morphological data of the existing target into the database; S202 summarizing and classifying the morphological parameters of the existing target, and obtaining several target types of the existing target; S203 obtaining the morphological parameters of the target target and comparing them with the morphological data of each type of target type, and determining the target type of the target target. It is worth noting that the target types of existing targets are roughly divided into circular targets, square targets, human-shaped targets, etc. The scoring standards for each type of target are different, and even some heterosexual targets will be designed according to the needs of shooting training. During shooting training, the target target needs to be imaged in advance to determine the shape and scoring standards of the target target, which is extremely time-consuming. Therefore, it is necessary to summarize and classify the morphological parameters of existing targets to classify them into several target types. When there is a new target, the morphological parameters of the target target are compared with the morphological data of each target type and classified into one target type to avoid the input operation of the target target. At the same time, there are many types of existing targets, and it is extremely complicated to manually summarize and classify the existing targets. Therefore, artificial intelligence is used to perform morphological recognition training on existing targets. Artificial intelligence training specifically includes AI training or SVM training, which are both artificial intelligence operating systems. Figure 2 , which specifically include:

[0060] S301 sets the existing target within the field of view of the thermal imaging device;

[0061] S302 adjusts the placement position and placement angle of the existing target to obtain a video stream of the target target;

[0062] S303 extracts features from the video stream of the target through artificial intelligence and obtains feature data of the target, and then calculates the stability of feature values ​​of the feature data;

[0063] S304 integrates the characteristic data of the target into morphological data, and saves the morphological data into a database.

[0064] Specifically, feature data are mostly geometric parameters, size parameters and contrast between the target and the background, etc. Moreover, feature value stability calculation needs to be performed on the feature data to obtain stable and reliable feature data. This is because the target is moving, which makes the feature data extremely unstable. The placement distance, placement angle and background environment of the target all have a great impact on the feature data. Therefore, it is necessary to extract the feature data that is less affected and integrate this feature data into morphological data to improve the reliability and accuracy of feature data capture.

[0065] S103 obtains each abnormal point on the target where the temperature change is abnormal based on the temperature data of each point of the target. In this embodiment, when the target is shot, the bullet will form a bullet hole in the target, and the bullet hole will produce a high temperature area on the image of the thermal imaging device. The temperature change of this area is greater than that of other areas, and an abnormal point is formed in the image of the thermal imaging device. The determination of the abnormal point mainly includes: S401 obtaining the maximum temperature of the background image and filling the background image with the maximum temperature; S402 obtaining the temperature data of each point of the target at a certain time point and the zero input response value and zero state response value of the target, and calculating the predicted change value of each point of the target; S403 obtaining the temperature data of each point of the target at the next time point, and calculating the actual change value of each point of the target; S404 comparing the predicted change value of each point of the target with the actual change value, if the actual change value of this point of the target is much greater than the predicted change value of the target, then this point is an abnormal point.

[0066] Specifically, filling the background image with the maximum temperature of the background image is to initialize the overall background image to avoid chaotic temperature changes, which is an initial operation. Both the zero input response value and the zero state response value are affected by the material, color and temperature of the target. Through the zero input response value and the zero state response value of the target, the predicted change value of each point of the target can be calculated, and then the temperature data of each point of the target at the next time point can be obtained to calculate the actual change value of each point of the target; finally, the two can be compared to determine whether this point is an abnormal point. Normally, the actual change value must be more than 5 times the predicted change value to be identified as an abnormal point.

[0067] Zero input response value refers to the temperature change law of an object under the assumption that the object has no external energy. Because when the temperature of an object is higher than absolute zero (0K = -273 degrees Celsius), the object will dissipate energy outward through heat conduction and heat radiation, which causes the temperature of the object itself to drop. The temperature of the object drops according to the following law. The zero input response value of the target target ΔT = (T b +273)×e -αt . ΔT is the temperature drop of an object in degrees Celsius; T b Refers to the temperature of the object, in degrees Celsius; α is the temperature change coefficient, which is determined by the heat capacity of the object and the heat conduction efficiency of the environment; t refers to the elapsed time, in ms, and e is the base of the natural exponent. The α value can be inferred from the temperature change at the bullet point. Due to the friction between the bullet and the target and the heat conduction of the high temperature on the bullet surface, the surface temperature of the object at the point where the target is hit increases significantly. After a few seconds (usually about 10 seconds), the temperature at this point can be greatly reduced. For the object at this point, the value of its zero-state response is consistent with that of other places on the target; due to the high temperature at the impact point, the zero input response value increases. Then the α value can be calculated.

[0068] The zero-state response value of the target means that objects in an environment above -273 degrees Celsius obtain energy from the environment through heat conduction and heat radiation, causing their own temperature to rise. The temperature of the object rises according to the following rules. The zero-state response value of the target ΔT = (T c +273)×e -αt +d. ΔT is the temperature rise of an object in degrees Celsius; T c Refers to the ambient temperature in degrees Celsius; α is the temperature variation coefficient, which is determined by the heat capacity of the object and the heat conduction efficiency of the environment, and the value is consistent with the zero input response value; d is the temperature rise caused by the energy obtained from the infrared light in the sunlight, and this value is affected by the heat capacity, surface area and surface color of the object; the existence of parameter d makes different objects present different temperatures in the sun. The d value is derived from the temperature and temperature difference of the target and the environment. t refers to the elapsed time in ms, and e is the base of the natural exponent. The ΔT values ​​of the respective zero input response values ​​and zero state response values ​​are calculated based on the same conditions. If the zero input response value of the target target is equal to the zero state response value of the target target, the object maintains thermal equilibrium and the object temperature does not change. Therefore, refer to Figure 3The zero-state response value and zero-input response value of the target can be obtained by conducting experiments in advance, that is, filling the background image area with the maximum temperature of the background image area, and the background image is darkened exponentially for many times, and then the background image is adjusted in combination with the video stream of the thermal imaging device to obtain the final background image, and the zero-state response value and zero-input response value are calculated in the process. Of course, the zero-state response value and zero-input response value can also be obtained by timely calculating the temperature values ​​of each point of the target, the ambient temperature and other parameters on the spot during the target training.

[0069] It is worth noting that it is also possible to determine whether it is an abnormal point through the temperature data of each point of the target at a certain time point and the temperature data of each point of the target at the next time point.

[0070] S104 removes and merges all abnormal points to obtain a bullet point area that conforms to the shooting shape. In this embodiment, all abnormal points are removed and merged to obtain a bullet point area that conforms to the shooting shape, thereby avoiding erroneous capture of the bullet point area and reading an erroneous shooting score. Removing and merging all abnormal points to obtain a bullet point area that conforms to the shooting shape specifically includes: S601 tracking and locating abnormal points in the time dimension and the space dimension; S602 removing abnormal points that are isolated and distributed in the space dimension; S603 removing abnormal points that appear briefly in the time dimension; S604 merging several abnormal points that are correlated in space and time into a target area; S605 performing morphological analysis on the target area, and the target area that conforms to the shooting shape is the bullet point area. Specifically, refer to Figure 4 The main reasons for abnormal points include the internal noise of the thermal imaging device, "hot spots" or "cold spots" with abnormal heat on the image, and foreign objects on the target. The characteristics of abnormal points caused by the internal noise of the thermal imaging device are isolated points and the randomness of the position of occurrence. Therefore, such abnormal points can be eliminated using the spatial position correlation algorithm. The abnormal points caused by the "hot spots" or "cold spots" with abnormal heat on the image mostly appear briefly, and can be eliminated using the time axis correlation algorithm. Foreign objects on the target cannot be distinguished well. Therefore, by merging several abnormal points into a target area, the morphology of the target area is analyzed, and the target area that does not conform to the shooting shape is eliminated.

[0071] S105 obtains the shooting score of the target according to the target type of the target and the position of the bullet point area in the target.

[0072] Specifically, because the target is lifted, fallen and turned over, the tracking and positioning of the abnormal points need to be combined with the tracking of the target, and then the displacement law of the bullet point area is calculated to achieve the correct positioning, prediction and tracking of the bullet point area.

[0073] Furthermore, the present application provides a multi-target thermal imaging capture system, using the multi-target thermal imaging capture method in any of the above embodiments, the multi-target thermal imaging capture system includes a thermal imaging device, an analysis module, a first calculation module, a screening module and a second calculation module, the thermal imaging device can identify and track multiple target targets; the analysis module can analyze the morphological parameters of the target target and obtain the target type of the target target; the first calculation module can calculate the temperature change data of each point on the target target and obtain the abnormal points on the target target; the screening module can screen each abnormal point on the target target and obtain the bullet point area; the second calculation module can calculate the shooting score of the bullet point area on the target target. In this embodiment, the multi-target thermal imaging capture system can automatically capture and track multiple target targets, dynamically screen and clean up interfering abnormal points, and read the shooting score on the target target more accurately and timely.

[0074] Furthermore, it also includes a training module, which can perform morphological recognition training on existing targets and classify the existing targets into several target types.

[0075] In this embodiment, the training module leverages the learning ability of artificial intelligence to enable it to perform morphological recognition training on existing targets and classify existing targets into several target types, so that the multi-target thermal imaging capture system can automatically classify the target targets, reduce the manual input of target types, and reduce manual operation steps.

[0076] It should be noted that the above embodiments can be freely combined as needed. The above are only preferred embodiments of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A multi-target thermal imaging capture method, characterized in that: include: Capturing and tracking multiple targets by thermal imaging equipment to obtain morphological parameters of the target and temperature data of each point; Determining the target type of the target according to the morphological parameters of the target; According to the temperature data of each point of the target, each abnormal point with abnormal temperature change on the target is obtained; All the abnormal points are removed and merged to obtain a bullet point area that conforms to the shooting shape; The shooting score of the target target is obtained according to the target type of the target target and the position of the bullet point area on the target target.

2. A multi-target thermal imaging capture method according to claim 1, characterized in that: Determining the target type of the target according to the morphological parameters of the target specifically includes: Use artificial intelligence to train the morphological recognition of existing targets and put the morphological data of existing targets into the database; Summarize and classify the morphological parameters of existing targets and obtain several target types of existing targets; The morphological parameters of the target are obtained and compared with the morphological data of each type of target of the existing targets, and the target type of the target is determined.

3. A multi-target thermal imaging capture method according to claim 3, characterized in that: The morphological recognition training of existing targets through artificial intelligence specifically includes: Set the existing target within the field of view of the thermal imaging device; Adjust the placement position and placement angle of the existing target to obtain the video stream of the existing target; Through artificial intelligence, feature extraction is performed on the video stream of the existing target to obtain feature data of the existing target, and then feature value stability calculation is performed on the feature data; The feature data of existing targets are integrated into morphological data, and the morphological data is saved in the database.

4. A multi-target thermal imaging capture method according to claim 1, characterized in that: According to the temperature data of each point of the target, obtaining each abnormal point of the target where the temperature changes abnormally includes: Obtaining the maximum temperature of the background image and filling the background image with the maximum temperature; Acquire the temperature data of each point of the target at a certain time point and the zero input response value and the zero state response value of the target, and calculate the predicted change value of each point of the target; Acquire the temperature data of each point of the target at the next time point, and calculate the actual change value of each point of the target; The predicted change value and the actual change value of each point of the target target are compared. If the actual change value of this point of the target target is much larger than the predicted change value of the target target, then this point is an abnormal point.

5. A multi-target thermal imaging bullet capture method according to claim 4, characterized in that: The zero input response value of the target ΔT=(T b +273)×e -αt , T b is the object temperature, α is the temperature change coefficient, t is the elapsed time, and e is the base of the natural exponent; The zero-state response value of the target ΔT=(T c +273)×e -αt +d,T c is the ambient temperature, α is the temperature variation coefficient, t is the elapsed time, d is the temperature rise caused by the energy obtained from the infrared light in the sunlight, and e is the base of the natural index.

6. A multi-target thermal imaging bullet capture method according to claim 1, characterized in that: All the abnormal points are removed and merged to obtain the bullet point area that conforms to the shooting shape, which specifically includes: Tracking and locating the abnormal point in time and space dimensions; Removing the abnormal points that are isolatedly distributed in the spatial dimension; Removing the abnormal points that appear briefly in the time dimension; Merging a plurality of the abnormal points that are correlated in space and time into a target area; The target area is subjected to a morphological analysis, and the target area that conforms to the shooting shape is the bullet point area.

7. A multi-target thermal imaging capture method according to claim 6, characterized in that: The tracking and positioning of the abnormal point needs to be combined with the tracking of the target, and then the displacement law of the bullet point area position is calculated to achieve the correct positioning, prediction and tracking of the bullet point area.

8. The multi-target thermal imaging capture method according to claim 1, characterized in that: The thermal imaging device distinguishes the target from the background according to the difference in texture or temperature between the target and the external environment.

9. A multi-target thermal imaging bullet capture system, characterized in that: The multi-target thermal imaging bullet capturing method according to any one of claims 1 to 8 comprises: A thermal imaging device capable of identifying and tracking multiple targets; An analysis module, which can analyze the morphological parameters of the target and obtain the target type of the target; A first calculation module, wherein the first calculation module is capable of calculating the temperature change data of each point on the target and obtaining the abnormal point on the target; A screening module, wherein the screening module can screen each abnormal point on the target and obtain a bullet point area; A second calculation module, wherein the second calculation module is capable of calculating the shooting score of the bullet point area on the target.

10. The multi-target thermal imaging bullet catching system according to claim 9, characterized in that: The invention also comprises a training module, which can perform morphological recognition training on existing targets and classify the existing targets into several target types.

Images