Infrared image interpretation method for engine state under complex working conditions

Abstract

The invention relates to a complex working condition engine state infrared image interpretation method, and relates to the field of engine test.The method comprises the steps that projection positions and envelope shapes of engine tail flames are obtained through calculation, and target frames are set to select the tail flames of all engines based on projection position information and envelope shape information; a background frame is selected according to the relation between centroid positions of all engine tail flame envelope shapes; calculating an average value, a median and a standard deviation of all pixel gray values in the target frame, calculating average values of all pixel gray values in the background frame and the whole image, and selecting a minimum value in the average values of all pixel gray values in all the target frames; selecting the average value of the gray values of all the pixels in all the background frames and the maximum value of the average values of the gray values of all the pixels in the whole image, and calculating the difference value between the minimum value and the maximum value; by comparing the difference value with the set threshold value, the on-off state of the engine can be judged. The method has the advantage of automatically judging the on-off state of the engine.

Description

Technical Field

[0001] The invention relates to the technical field of engine testing, and in particular to a method for interpreting infrared images of engine states in complex working conditions. Background Art

[0002] Engine tail flame measurement is a common measurement scenario in infrared testing, and the judgment of the engine working status (such as on and off) is an important work link closely related to the infrared characteristics of the engine tail flame and engine performance analysis. Considering the measurement efficiency and test cost, under the premise of reasonable and safe spatial arrangement, infrared testing will include multiple engines under test in the same measurement field of view and carry out synchronous measurement of multiple engine tail flames. After the test, the infrared scene measurement image obtained by the test is interpreted, and the engine is judged whether it is turned on based on the change in grayscale value.

[0003] However, since the working states of multiple engines at the same time are arranged and combined, multiple measured states will be formed. Even when some high-power engines are turned on, it will affect the judgment of the switch states of other engines. Therefore, manual judgment is usually used to judge the switch states of engines in such complex working conditions. However, although the manual judgment method has high accuracy, it is limited by the recognition frequency of the human eye, and the judgment efficiency is extremely low, which is difficult to meet the needs of today's high frame rate and ultra-large data volume infrared measurement data processing. In addition, the judgment criteria that vary from person to person, operator fatigue and training costs are also disadvantages of the manual judgment method.

[0004] Therefore, in view of the above shortcomings, it is necessary to provide an infrared image interpretation method for engine status under complex working conditions. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] The technical problem to be solved by the present invention is to solve the problem of lack of effective automatic judgment method for engine switch status due to mutual interference of multiple engines in the same field of view in the engine tail flame measurement scene.

[0007] (II) Technical solution

[0008] In order to solve the above technical problems, the present invention provides a method for interpreting infrared images of engine status under complex working conditions, comprising the following steps:

[0009] Ⅰ. According to the distribution position of each engine and the basic shape characteristics of the tail flame, the projection position and envelope shape of the engine tail flame on the measurement receiving plane are calculated;

[0010] II. Based on the projection position information and envelope shape information, set the target box to select the tail flame of each engine;

[0011] III. Select the background frame according to the relationship between the centroid positions of the tail flame envelopes of each engine, and the background frame and the target frame have the same size;

[0012] IV. Calculate the average, median and standard deviation of the grayscale values ​​of all pixels in the target frame, calculate the average grayscale values ​​of all pixels in the background frame, and calculate the average grayscale values ​​of all pixels in the entire image;

[0013] V. Select the minimum value among the average values ​​of the grayscale values ​​of all pixels in all target frames, select the maximum value among the average values ​​of the grayscale values ​​of all pixels in all background frames and the average values ​​of the grayscale values ​​of all pixels in the entire image, and calculate the difference between the minimum value and the maximum value;

[0014] VI. By comparing the difference with the set threshold, the engine on or off status can be determined.

[0015] As a further illustration of the present invention, preferably, the relative position relationship between the engine to be analyzed and the high-power engine is determined by acquiring engine deployment position information.

[0016] As a further illustration of the present invention, preferably, both the target frame and the background frame are selected as rectangles.

[0017] As a further illustration of the present invention, preferably, the centroid of one of the background frames of the small engine is located at a distance from half to a quarter of the line connecting the centroid of the tail flame envelope shape of the large engine and the centroid of the tail flame envelope shape of the engine.

[0018] As a further illustration of the present invention, preferably, the centroid position of another background frame of the small engine is located at a distance from half to one quarter of the outer extension line of the centroid of the tail flame envelope shape of the large engine and the centroid of the tail flame envelope shape of another small engine.

[0019] As a further illustration of the present invention, preferably, the background frame does not overlap with the target frame position.

[0020] As a further illustration of the present invention, preferably, the threshold value is x times the standard deviation of the grayscale values ​​of all pixels in the target frame of the desired engine state.

[0021] As a further illustration of the present invention, preferably, the initial value of x is 1.

[0022] As a further illustration of the present invention, preferably, when the difference between the minimum value and the maximum value is greater than a threshold value, the measured engine is in an on state, and if it is less than the threshold value, the measured engine is in an off state.

[0023] (III) Beneficial effects

[0024] The above technical solution of the present invention has the following advantages:

[0025] The present invention starts from the information of the engine deployment position, combines the geometric characteristics of the engine tail flame, and uses infrared image processing difference technology to select the tail flame area and optimize the background area respectively to calculate statistical data. After setting the threshold iterative detection, the automatic judgment and result output of the engine switch state are realized, which greatly improves the efficiency of infrared image interpretation of the engine state under complex working conditions, and avoids the data processing that varies from person to person and the misjudgment of the state caused by personnel fatigue and training effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a state diagram of the engine A and the engine B of the present invention working, and the engine C not working;

[0027] Figure 2 This is a state diagram of the engine A of the present invention being in operation, and the engines B and C being inoperative;

[0028] Figure 3 This is a state diagram of the engine B of the present invention in operation, and the engines A and C inoperative;

[0029] Figure 4 This is a diagram of the non-operating state of engines A, B and C of the present invention;

[0030] Figure 5 It is a working state diagram of engines A, B and C of the present invention;

[0031] Figure 6 This is a state diagram of the engine A and C of the present invention working and the engine B not working;

[0032] Figure 7 This is a state diagram of the present invention in which engines B and C are working and engine A is not working;

[0033] Figure 8 This is a diagram of the working state of the engine A when the engines A and B of the present invention are not working;

[0034] Fig. 9 It is a state diagram of the tail flame ejection of engine A and engine B of the present invention;

[0035] Fig.10 This is a diagram of the normal tail flame ejection state of engines A and C of the present invention;

[0036] Fig.11 This is a diagram showing that the tail flame of engine A of the present invention is in a weak state;

[0037] Fig.12 It is a state diagram of the background frame area affected by the tail flame of engine C of the present invention;

[0038] Fig.13 This is a state diagram of the target frame area affected by the tail flame of engine C jet of the present invention.

[0039] In the figure: 1. Engine A; 2. Engine B; 3. Tail flame A; 4. Tail flame B; 5. Engine C; 6. Tail flame C; 7. Second background frame A; 8. Target frame A; 9. First background frame A; 10. First background frame B; 11. Target frame B; 12. Second background frame B. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] A method for interpreting infrared images of engine status under complex working conditions comprises the following steps:

[0042] Ⅰ. Communicate with on-site testers, query test reports, obtain the deployment position information of multiple engines, and determine the relative position relationship between the engine to be analyzed and the high-power engine by obtaining the engine deployment position information.

[0043] Ⅱ. Based on the projection position information and envelope shape information, set the target box to select the tail flame of each engine.

[0044] III. Select the background frame according to the relationship between the centroid positions of the tail flame envelope shapes of each engine. The background frame and the target frame have the same size.

[0045] IV. Calculate the average, median and standard deviation of the grayscale values ​​of all pixels in the target frame, calculate the average grayscale values ​​of all pixels in the background frame, and calculate the average grayscale values ​​of all pixels in the entire image.

[0046] V. Select the minimum value among the average values ​​of the grayscale values ​​of all pixels in all target frames, select the maximum value among the average values ​​of the grayscale values ​​of all pixels in all background frames and the average values ​​of the grayscale values ​​of all pixels in the entire image, and calculate the difference between the minimum value and the maximum value.

[0047] VI. By comparing the difference with the set threshold, the engine on or off status can be determined.

[0048] Combination Figure 1-Figure 8The present invention takes three engines as an example, wherein Engine A1 and Engine B2 are engines with relatively low power, and Engine C is a high-power engine. According to the requirements of the engine test demander, Engine A, Engine B and Engine C are at the same test site, wherein Engine A and Engine B are engines for which the infrared radiation characteristics of tail flames need to be analyzed, and the tail flames of the two are ejected in opposite directions and are parallel to the observation plane. Engine C is a high-power engine that will cause interference, and the tail flame ejection direction is perpendicular to the observation plane.

[0049] By communicating with the personnel of the engine unit and the on-site test personnel, we learned that the tail flame shapes of Engine A, Engine B and Engine C are all approximately ellipsoids, and different working conditions will affect the length of the major axis of the ellipsoid. Therefore, a rectangular frame can be used to select the target. When the field of view of the measuring device, the distance between the measuring device and the engine, and the tail flame length of the rated power engine are known, combined with the number of pixels, pixel size and focal length of the measuring device, the center pixel position of the rectangular frame of the engine tail flame and the number of pixels of the rectangular side length are preliminarily estimated to obtain the projection position information and envelope shape information.

[0050] The measured data of the engine tail flame is parsed and read to obtain the image frame sequence number and time axis information, and the image frame sequence number is used as the unique ID of the data processing result. Figures 1 to 8 There are eight measurement conditions. Depending on whether they are affected by the non-tested high-power engine C, these eight conditions can be divided into two categories:

[0051] 1. If Figure 1-Figure 4 is not affected by Engine C.

[0052] 2. If Figure 5-Figure 8 is affected by Engine C.

[0053] like Figure 9-13 , using the calculated projection position and envelope shape information, a rectangular frame is used as the target frame selection shape, and in the infrared measurement image, the four endpoints of the rectangular frames of target frame A 8 and target frame B 11 are selected respectively to form target frame A 8 and target frame B 11 for selecting tail flame A 3 and tail flame B 4 respectively.

[0054] Using the relative positional relationship between engine A1 and engine B2 and engine C3, according to the relationship between the centroid positions of the engine tail flame envelope shape, two background frames are selected for engine A1 and engine B2 respectively, and the background frame has the same size as the target frame. Among them, the centroid position of the first background frame A9 is located at a position close to the tail flame A3 in the line connecting the centroid of the tail flame C6 envelope shape and the centroid of the tail flame A3 envelope shape. The centroid position of the second background frame A7 is located at a position close to the tail flame C6 in the outer extension line of the centroid of the tail flame B4 envelope shape and the centroid of the tail flame C6 envelope shape. The centroid position of the first background frame B10 is located at a position close to the tail flame B4 in the line connecting the centroid of the tail flame C6 envelope shape and the centroid of the tail flame B4 envelope shape. The centroid position of the second background frame B12 is located at a position close to the tail flame C6 in the outer extension line of the centroid of the tail flame A3 envelope shape and the centroid of the tail flame C6 envelope shape. In addition, the target frame does not overlap with any background frame.

[0055] Calculate the average DN1, median DN2 and standard deviation std of all pixel grayscale values ​​in target frame A8 or target frame B11, calculate the average DN3 of all pixel grayscale values ​​in first background frame A9 or first background frame B10, calculate the average DN4 of all pixel grayscale values ​​in second background frame A7 or second background frame B12, and calculate the average DN5 of all pixel grayscale values ​​in the entire image. Select the minimum DNt of DN1 and DN2, select the maximum DNb of DN3, DN4 and DN5, and calculate the difference dDN between DNt and DNb. Set x times std as the threshold, where the initial value of x is 1, and the value of x can be adjusted through small sample testing.

[0056] When dDN is greater than x times std, engine A1 or engine B2 is in the on state, otherwise it is in the off state. The output method of the state determination result of engine A1 and engine B2 is not limited.

[0057] In summary, the above method is used to judge the state of the engine A1 or engine B2 to be tested, realizes the automatic judgment and result output of the engine switch state, greatly improves the efficiency of infrared image interpretation of the engine state under complex working conditions, and avoids the misjudgment of the state caused by data processing that varies from person to person and personnel fatigue and training effects. Moreover, the confidence is weakened only when the background frame area is affected after the engine C5 is turned on and the tail flame A3 or tail flame B4 is weak, so the application range is wider under the premise of high judgment accuracy.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for interpreting infrared images of engine status under complex working conditions. Features: The following steps are included: Ⅰ. According to the distribution position of each engine and the basic shape characteristics of the tail flame, the projection position and envelope shape of the engine tail flame on the measurement receiving plane are calculated; II. Based on the projection position information and envelope shape information, set the target box to select the tail flame of each engine; III. Select the background frame according to the relationship between the centroid positions of the tail flame envelopes of each engine, and the background frame and the target frame have the same size; IV. Calculate the average, median and standard deviation of the grayscale values ​​of all pixels in the target frame, calculate the average grayscale values ​​of all pixels in the background frame, and calculate the average grayscale values ​​of all pixels in the entire image; V. Select the minimum value among the average values ​​of the grayscale values ​​of all pixels in all target frames, select the maximum value among the average values ​​of the grayscale values ​​of all pixels in all background frames and the average values ​​of the grayscale values ​​of all pixels in the entire image, and calculate the difference between the minimum value and the maximum value; VI. By comparing the difference with the set threshold, the engine on or off status can be determined.

2. According to claim 1, a method for interpreting infrared images of engine status under complex working conditions, Features: The relative position relationship between the engine to be analyzed and the high-power engine is determined by acquiring the engine deployment position information.

3. According to claim 2, a method for interpreting infrared images of complex engine conditions, Features: Both the target frame and the background frame are selected as rectangles.

4. The infrared image interpretation method for complex engine conditions according to claim 3, Features: The centroid of one of the background frames of the small engine is located at a position close to half to a quarter of the latter of the line connecting the centroid of the tail flame envelope shape of the large engine and the centroid of the tail flame envelope shape of the engine.

5. The infrared image interpretation method of a complex engine condition according to claim 4, Features: The centroid of another background frame of the small engine is located at a position close to half to one quarter of the outer extension line of the centroid of the tail flame envelope shape of the large engine and the centroid of the tail flame envelope shape of another small engine.

6. The infrared image interpretation method for complex engine conditions according to claim 5, Features: The background frame does not overlap with the target frame position.

7. The infrared image interpretation method for complex engine conditions according to claim 1, Features: The threshold is x times the standard deviation of the grayscale values ​​of all pixels in the target box of the desired engine state.

8. The infrared image interpretation method for complex engine conditions according to claim 7, Features: The initial value of x is 1.

9. The infrared image interpretation method for complex engine conditions according to claim 8, Features: When the difference between the minimum value and the maximum value is greater than a threshold value, the measured engine is in an on state; if it is less than the threshold value, the measured engine is in an off state.

Images