Signal state discrimination system for precision approach path indicator and control method thereof

Abstract

Disclosed are a signal state discrimination system of a precision approach path indicator and a control method thereof. The signal state discrimination system of the precision approach path indicator includes a display, a camera unit that acquires an image including the precision approach path indicator, and a processor that performs image preprocessing for object recognition of the precision approach path indicator according to the image, extracts a red light region by performing RGB masking on the precision approach path indicator recognized as an object, and applies image binarization and image contouring to calculate the number of the extracted red light region and display the same on the display. Thus, without relying on the human eye, the automatic method can be used to discriminate the glide path information, thereby ensuring reliability and consistency, and stable discrimination can be achieved under various light intensities and weather conditions, thereby minimizing the risk of accidents that can occur during flight landing.

Description

Technical Field

[0001] This invention relates to a signal status discrimination system and control method for a precision approach path indicator, and more specifically, to a signal status discrimination system and control method for a precision approach path indicator used to analyze the light intensity position of the precision approach path indicator to determine its signal status.

[0002] This invention was developed by the Electronics Technology Research Institute of Korea (ETRI) with the support of the Ministry of Science and ICT, for the research project "Development of SW Computing Industry Resource Technology", with the research topic titled "Development and Verification of Airborne Intelligent Avionics SW Platform Technology for Identification and Response to Abnormal Urban Air Traffic Situations" (Project No.: 00400712, Project Unique Number: 2710008225). Background Technology

[0003] A Precision Approach Path Indicator (PAPI) is a visual guidance device installed beside the runway to help maintain an accurate glide angle during aircraft landing. It provides a combination of red and white lighting so that the pilot can immediately identify with the naked eye whether they are approaching the correct glide path.

[0004] This precision approach path indicator consists of four lights, each with a different angle of illumination. Therefore, during flight, the pilot can determine the current glide path status by observing the color combinations seen on the precision approach path indicator.

[0005] Existing precision approach path indicator systems rely on pilots visually confirming signal status. This method has several problems: recognition errors can occur due to ambient lighting conditions, weather, and the position of sunlight; additionally, it depends on the pilot's subjective judgment, which may lead to pilot misunderstanding of signals; and it is difficult to apply to automated systems such as drones or UAM (Urban Air Traffic Assistance).

[0006] As a result, there is an increasing demand for the development of technologies that can determine signal status by analyzing the light intensity position of precision approach path indicators in real time. Summary of the Invention

[0007] Technical problems to be solved

[0008] The purpose of this invention is to provide a signal status determination system and control method for a precision approach path indicator (PMI) for analyzing the light position of the PMI to determine the signal status.

[0009] means of solving technical problems

[0010] A signal status discrimination system for a precision approach channel indicator according to an embodiment of the present invention, for achieving this purpose, may include: a display; a camera unit that acquires an image including the precision approach channel indicator; and a processor that performs image preprocessing to perform object recognition on the precision approach channel indicator based on the image, extracts red light regions by performing RGB masking on the precision approach channel indicator for the object recognition, and applies image binarization and image contouring to calculate the number of the extracted red light regions and display them on the display.

[0011] The processor can use an object detection algorithm to identify objects in the precision approach path indicator and set them as regions of interest.

[0012] Furthermore, the processor can adjust the threshold of the RGB values ​​for the region of interest and extract the red light area from the precision approach path indicator identified by the object.

[0013] Furthermore, the processor can apply image binarization to the precision approach path indicator identified by the object with the threshold of the RGB values ​​adjusted, thereby displaying the extracted red light area as white and the remaining area as black.

[0014] Furthermore, the processor can apply image contours to the object recognition precision approach path indicator identified by the above-mentioned image binarization, and extract the outer contour lines of the area displayed in white.

[0015] In addition, the processor can count the number of extracted outer contour lines to calculate the number of extracted red light areas.

[0016] In addition, the processor can control the display to show the signal status corresponding to the number of red light areas extracted above.

[0017] On the other hand, a control method for a signal status discrimination system for a precision approach channel indicator according to an embodiment of the present invention may include: a step of acquiring an image including the precision approach channel indicator; a step of performing image preprocessing to identify objects in the precision approach channel indicator based on the image; a step of extracting red light regions by applying an RGB mask to the precision approach channel indicator identified by the object; and a step of calculating and displaying the number of extracted red light regions by applying image binarization and image contouring.

[0018] In the above image preprocessing step, an object detection algorithm can be used to identify objects in the precision approach path indicator and set them as regions of interest.

[0019] Furthermore, in the step of extracting the red light area, the threshold of the RGB value of the region of interest can be adjusted to extract the red light area from the precision approach path indicator identified by the object.

[0020] Furthermore, in the step of calculating and displaying the number of extracted red light areas, image binarization can be applied to the precision approach channel indicator for object recognition with the threshold of the RGB values ​​adjusted, so that the extracted red light areas are displayed as white and the remaining areas are displayed as black. Image contouring is applied to the precision approach channel indicator for object recognition with image binarization applied, the outer contour lines of the areas displayed as white are extracted, and the number of the extracted outer contour lines is counted to calculate the number of extracted red light areas.

[0021] On the other hand, a computer-readable storage medium according to an embodiment of the present invention may include a program for executing a control method for a signal status determination system of a precision approach path indicator in a computer.

[0022] Invention Effects

[0023] According to the various embodiments of the present invention described above, glide path information can be determined automatically without relying on the human eye, thereby ensuring reliability and consistency. Stable determination can also be achieved under different lighting and weather conditions, thereby minimizing the risk of accidents that may occur during flight and landing. Attached Figure Description

[0024] Figure 1 This is a diagram illustrating the configuration of a signal status determination system for a precision approach path indicator according to an embodiment of the present invention.

[0025] Figure 2 This is a diagram illustrating a precision approach path indicator according to an embodiment of the present invention.

[0026] Figure 3 This is a diagram illustrating a region of interest obtained through image preprocessing according to an embodiment of the present invention.

[0027] Figure 4 This is a table used to explain the status information provided by the hue combination based on a precision approach path indicator according to an embodiment of the present invention.

[0028] Figure 5 This is a diagram illustrating an RGB mask according to an embodiment of the present invention.

[0029] Figure 6 This is a diagram used to illustrate image binarization and image contours according to an embodiment of the present invention.

[0030] Figure 7 This is a graph showing the number of red light regions extracted according to an embodiment of the present invention.

[0031] Figure 8 This is a flowchart illustrating a control method for a signal status determination system of a precision approach path indicator according to an embodiment of the present invention.

[0032] Figure 9 It is shown Figure 1 The diagram shows the specific structure of the signal status determination system for a precision approach path indicator.

[0033] Figure 10 This is a diagram of a software module stored in a storage unit according to an embodiment of the present invention.

[0034] Explanation of reference numerals in the attached figures

[0035] 100: Signal Status Discrimination System for Precision Approach Course Indicators

[0036] 110: Monitor; 120: Camera Unit

[0037] 130: Processor; 140: Storage Unit Detailed Implementation

[0038] The present invention will now be described in further detail with reference to the accompanying drawings. Furthermore, in the process of describing the present invention, detailed descriptions of related well-known functions or configurations that are deemed likely to obscure the spirit of the invention will be omitted. Additionally, the terminology used herein is defined in light of the functions employed in the invention and may vary depending on the intent or relationship of the user or operator. Therefore, it should be defined based on the overall content of this specification.

[0039] Figure 1 This is a diagram illustrating the configuration of a signal status determination system for a precision approach path indicator according to an embodiment of the present invention.

[0040] Reference Figure 1 According to an embodiment of the present invention, a signal status determination system 100 for a precision approach path indicator may include a display 110, a camera unit 120, and a processor 130.

[0041] Among them, the signal status discrimination system 100 of the precision approach channel indicator is a system for accurately judging the signal status based on the light quantity and position of the precision approach channel indicator. This system can be implemented by a system composed of separate devices, by a single device, or by a software module embedded in the device.

[0042] For example, the signal status discrimination system 100 of the precision approach channel indicator according to an embodiment of the present invention can also be implemented by software embedded in a vision camera, thereby performing all processing of image preprocessing, RGB masking, image binarization and image contouring by performing object recognition on the precision approach channel indicator through the vision camera.

[0043] The display 110 is a component or device that includes a screen capable of displaying images, and can be implemented through the screen of LCD, LED, imaging device, screen panel, electronic whiteboard, video wall, PDA, tablet device, desktop computer, laptop computer, user terminal device, etc.

[0044] Furthermore, the camera unit 120 can acquire images including a precision approach path indicator, as described above, which can be achieved through a vision camera.

[0045] The camera unit 120 can capture images, including those from a precision approach path indicator, in real time and continuously during a predetermined period of time. For example, it can be installed inside or outside the aircraft, and multiple camera units 120 can be installed.

[0046] In addition, the processor 130 performs image preprocessing based on the image acquired by the camera unit 120 to perform object recognition on the precision approach channel indicator, performs RGB masking on the precision approach channel indicator to extract the red light area, applies image binarization and image contouring, calculates the number of extracted red light areas and displays them on the display 110.

[0047] Specifically, after the processor 130 performs object recognition and extraction on the precision approach path indicator from the image acquired by the camera unit 120, it extracts the area corresponding to the red light source from among the multiple light sources contained in the precision approach path indicator, and performs data processing on this, thereby calculating the number of areas corresponding to the red light source.

[0048] In addition, the processor 130 determines the signal status corresponding to the number of areas of the calculated red light source and can notify the pilot or user through the display 110.

[0049] Therefore, the signal status discrimination system 100 for a precision approach channel indicator according to an embodiment of the present invention extracts only the precision approach channel indicator from the acquired image, extracts only the light source of a specific hue region contained in the light source of the precision approach channel indicator and binarizes it, thereby enabling the determination of the number of light sources of the specific hue region required for signal status discrimination, and improving the accuracy of signal status discrimination.

[0050] Specifically, the processor 130 can use an object detection algorithm to identify objects in the precision approach path indicator and set them as regions of interest.

[0051] Figure 2 This is a diagram illustrating a precision approach path indicator according to an embodiment of the present invention.

[0052] Reference Figure 2 The image shown is obtained by the camera unit 120 according to an embodiment of the present invention, and it can be seen that the precision approach path indicator contained in the image is identified by the red box.

[0053] That is, the processor 130 does not perform image processing on the entire image acquired by the camera unit 120, but only extracts the required area, namely the precision approach channel indicator area. For the corresponding image, an object detection algorithm can be used to find only the precision approach channel indicator area and display the position of the corresponding precision approach channel indicator area with a bounding box (red box).

[0054] Among the available object detection algorithms are R-CNN (Region-based Convolutional Neural Networks), Fast R-CNN, Faster R-CNN, YOLO (You Only Look Once), and SSD (Single Shot MultiBox Detector).

[0055] Additionally, the processor 130 can designate the precision approach path indicator identified by the object as the Region of Interest (ROI).

[0056] Figure 3 This is a diagram illustrating a region of interest obtained through image preprocessing according to an embodiment of the present invention.

[0057] Reference Figure 3 The example shown is an example of setting a precision approach path indicator for object identification as a region of interest 310, and the processor 130 can efficiently perform image or video processing only on the region of interest 310 set in this way.

[0058] For example, when applying RGB masking, image binarization, image contouring, and other processing methods described later, the processor 130 can process only the set region of interest 310 as an object, thereby speeding up the processing and enabling it to provide the pilot or user with accurate information about the descent path without delay.

[0059] on the other hand, Figure 4 This is a table used to explain the status information provided by the hue combination based on a precision approach path indicator according to an embodiment of the present invention.

[0060] Reference Figure 4 A precision approach path indicator typically consists of four light sources. By analyzing the color combinations presented by each of the four light sources, it is possible to determine the current state of the aircraft's glide path.

[0061] Specifically, when you see two red light sources and two white light sources, you can determine that the current glide angle of the aircraft is normal. When you see three red light sources and one white light source, you can determine that the current glide angle of the aircraft is too low. When you see only red light sources, you can determine that the current glide angle of the aircraft is very low.

[0062] Furthermore, when one red light source and three white light sources are observed, it can be determined that the current glide angle of the aircraft is too high. When all white light sources are observed, it can be determined that the current glide angle of the aircraft is extremely high.

[0063] In summary, when there are two red light sources, the glide angle of the current aircraft is a normal glide angle. With two as the baseline, the more red light sources there are, the lower the glide angle of the current aircraft can be judged. With two as the baseline, the fewer red light sources there are, the higher the glide angle of the current aircraft can be judged.

[0064] Therefore, as will be explained later, the processor 130 according to an embodiment of the present invention calculates the number of such red light sources and can determine the signal state based on this.

[0065] In addition, the processor 130 adjusts the threshold of the RGB values ​​for the region of interest 310, thereby extracting the red light area from the precision approach path indicator that identifies the object.

[0066] Figure 5 This is a diagram used to illustrate an RGB mask according to an embodiment of the present invention.

[0067] Reference Figure 5 The processor 130 can set a threshold for the RGB values, thereby accurately extracting the red light areas 510 and 520 from the precision approach path indicator for object recognition.

[0068] Here, RGB mask refers to setting a threshold for RGB values ​​and extracting the region within the corresponding threshold range.

[0069] Specifically, an image can typically be represented as an RGB (Red, Green, Blue) hue model, where the three hue values ​​(R, G, B) of each pixel are integers between 0 and 255. For example, white (R, G, B) can be represented as (255, 255, 255), and red (R, G, B) can be represented as (255, 0, 0).

[0070] In addition, in order to accurately separate or extract specific colors within an image, a threshold range of upper or lower limits for RGB values ​​can be set, allowing only pixels with (R, G, B) values ​​that fall within this threshold range to be extracted.

[0071] For example, red (R, G, B) can be represented in various ways such as (250, 20, 20) and (200, 10, 10). Therefore, it is necessary to specify which ranges are considered red. This range can be set by using the threshold values ​​of the RGB values ​​mentioned above.

[0072] That is, the processor 130 adjusts the threshold of RGB values ​​used to specify the range of what is considered a red area, and extracts pixels with RGB values ​​within the threshold range of the adjusted RGB values, so that, as described above, the red light area can be accurately extracted from the precision approach path indicator for object recognition.

[0073] As another embodiment, since RGB values ​​are sensitive to illumination, the processor 130 can also convert RGB values ​​to HSV (Hue, Saturation, Value) color space and set a threshold according to changes in illuminance.

[0074] Specifically, the HSV color space consists of three elements: hue, saturation, and value, and can be converted to RGB.

[0075] For example, if the RGB values ​​of a precision approach path indicator for object recognition are affected by excessive sunlight or sunset light, the color space can be converted to HSV color space and the threshold adjusted to extract the desired color region.

[0076] On the other hand, the processor 130 can apply image binarization to the precision approach path indicator identified by the object with the threshold of the RGB values ​​adjusted, and display the extracted red light area as white and the remaining area as black.

[0077] In addition, the processor 130 can extract the outer contour of the area displayed in white by applying image contours to the precise approach path indicator for object recognition using image binarization.

[0078] Figure 6This is a diagram used to illustrate image binarization and image contours according to an embodiment of the present invention.

[0079] Reference Figure 6 For the precision approach path indicator that is extracted from the red light area after adjusting the threshold of RGB values, image binarization is applied, so that the red light areas 510 and 520 are converted into white areas 610 and 620, and the remaining areas other than the red light areas 510 and 520 are converted into black areas 630.

[0080] Specifically, image binarization is a process that converts the original image into black and white. It compares the pixel value with a reference threshold; if the value is higher than the threshold, it is converted to white; if the value is lower than the threshold, it is converted to black.

[0081] This type of image binarization is mainly used in object extraction, boundary line extraction, contour analysis, OCR, etc.

[0082] The specific image binarization process is as follows: First, grayscale conversion is applied to convert the RGB image into a grayscale image. At this time, each pixel has a brightness value in the range of 0 to 255.

[0083] Then, using a user-specified threshold as a benchmark, the pixel values ​​are compared to extract bright objects as white and dark objects as black.

[0084] For example, when the processor 130 performs grayscale conversion on the precision approach path indicator identified by the object in the red light areas 510 and 520 to convert it into a grayscale image and sets the threshold to 127, if the pixel value of each pixel constituting the grayscale image is greater than 127, it can be converted to white (255), and if it is less than 127, it can be converted to black (0).

[0085] like Figure 6 As shown, based on the above image binarization processing results, the red light areas 510 and 520 are converted into white areas 610 and 620, and the remaining areas other than the red light areas 510 and 520 are converted into black areas 630.

[0086] On the other hand, in order to accurately apply this image binarization, the processor 130 can amplify the brightness value of a specific hue region before applying the image binarization.

[0087] For example, after adjusting the threshold of the RGB values ​​of the precision approach path indicator for object recognition and extracting the red light areas 510 and 520, the processor 130 can amplify the brightness values ​​of the pixels constituting the extracted red light areas 510 and 520.

[0088] In addition, the processor 130 applies image binarization while amplifying the brightness values ​​of the pixels constituting the extracted red light areas 510 and 520, thereby enabling the processor to accurately convert only the red light areas 510 and 520 into white areas 610 and 620.

[0089] Furthermore, the processor 130 applies image contours to the precision approach path indicator identified by the object, which is converted from red light areas 510 and 520 to white areas 610 and 620 and from areas other than red light areas 510 and 520 to black areas 630 after image binarization, thereby extracting the outer contour lines of the areas 610 and 620 that are displayed as white.

[0090] Image contour refers to the lines that connect the boundaries of an image with the same or similar brightness or hue values. It is a processing method for detecting the contour lines or boundaries of objects.

[0091] By applying this image contour for detection, objects can be identified and their quantity or location can be extracted based on the contour lines. This makes it suitable for object detection and enables numerical analysis of the object's width, length, area, position, etc. It can also be applied to computer vision in areas such as robot vision, autonomous driving, image recognition, and manufacturing inspection.

[0092] In addition, as described above, when the processor 130 extracts the outer contour lines of the regions 610 and 620 that are displayed in white using the image contour, it can count the number of extracted outer contour lines, that is, count the number of times the contour is applied.

[0093] Furthermore, the processor 130 can calculate the number of extracted red light areas 510 and 520 based on the number of counted outer contour lines. Figure 6 The number of extracted outer contour lines is two, so the processor 130 calculates that the number of extracted red light areas 510 and 520 is two.

[0094] Figure 7 This is a graph showing the number of extracted red light areas according to an embodiment of the present invention.

[0095] Reference Figure 7 The processor 130 calculates that there are two red light areas 510 and 520 extracted and displays "Detected lights: 2" on the display 110. The pilot or user can intuitively confirm that there are two red lights without visual confirmation and can immediately determine whether the current glide angle of the aircraft is normal.

[0096] In addition, the processor 130 can control the display 110 to display the signal status corresponding to the number of extracted red light areas 510 and 520, instead of simply calculating the number of extracted red light areas 510 and 520.

[0097] That is, such as Figure 4 As shown in the table, the processor 130 can directly display on the display 110 the signal status corresponding to the number of extracted red light areas 510, 520, for example, when Figure 6 When the number of outer contour lines in the middle count is two, the processor 130 determines that the number of red light areas is two, and displays the signal status corresponding to the number of red light areas being two, i.e., "normal glide angle", on the display 110 through text or images, instead of simply displaying two on the display 110.

[0098] In addition, the processor 130 can also output the number or signal status of the aforementioned red light areas via audio signals, and can also display them on the aircraft's HUD (Head-Up Display).

[0099] On the other hand, it is conceivable that there are multiple precision approach path indicators (PAP indicators) on the runway. In this case, the processor 130 can perform object recognition on the multiple PAP indicators based on the image acquired by the camera unit 120 and set them as regions of interest respectively. It then applies threshold adjustment of the RGB values, image binarization, and image contouring for each region of interest to calculate the number of red light areas in each PAP indicator. By comparing the calculated number of red light areas with each other, the accuracy can be verified and the signal status can be determined.

[0100] If the number of each red light zone is different, the processor 130 can automatically output a hazard notification message on the display 110, so that the pilot can know that the aircraft's glide angle is not within the normal range.

[0101] As described above, by processing multiple precision approach path indicators, it is possible to deal with unexpected situations such as malfunctions of a single precision approach path indicator or light interference caused by external light.

[0102] On the other hand, the processor 130 can automatically correct color distortion caused by weather conditions, time period, aging, etc., based on deep learning, thereby determining the signal status as described above.

[0103] For example, due to external factors such as weather conditions, sun position, time of day, lens contamination, and aging of the precision approach path indicator (PAP), the hue of the PAP may be captured differently or the hue of the PAP may change. In such cases, processing based on RGB value thresholds may result in errors. To prevent such errors, the processor 130 can perform automatic correction based on deep learning.

[0104] Specifically, the processor 130 acquires images from multiple precision approach path indicators (including various weather conditions such as sunny, cloudy, rainy, foggy, and nighttime), multiple time periods, and multiple locations via the camera unit 120, and assigns actual hue status information to each acquired image.

[0105] In addition, the processor 130 can set the captured image, illuminance value and RGB histogram information as input values, and set the RGB threshold range or RGB mask image in pixel units as output values ​​to generate an automatic correction deep learning model based on environmental information.

[0106] The processor 130 can learn the set input and output values ​​of the CNN-based Segmentation + MLP architecture, thereby generating an automatically corrected deep learning model based on environmental information.

[0107] In addition, the processor 130 can continuously learn images and metadata about environmental information (time period, weather, etc.) through the generated automatic correction deep learning model based on environmental information, and can analyze illuminance, contrast, RGB distribution, and classify the current environmental state.

[0108] In addition, the processor 130 is based on an environmental information-based automatic correction deep learning model that has learned about the current environmental state. It can calculate an appropriate RGB threshold or correction parameter, and based on the calculated RGB threshold or correction parameter, it can perform object recognition, region of interest setting, and RGB value threshold adjustment for the precision approach path indicator, thereby accurately extracting the red light area.

[0109] That is, the processor 130 can generate mask parameters (threshold of RGB values) corresponding to the current environmental state through the above-mentioned automatic correction deep learning model based on environmental information, and apply the mask parameters generated in this way to accurately extract the red light area without being affected by external environmental factors.

[0110] Figure 8 This is a flowchart illustrating a control method for a signal status determination system of a precision approach path indicator according to an embodiment of the present invention.

[0111] Reference Figure 8According to an embodiment of the present invention, the control method of the signal status discrimination system of the precision approach channel indicator may include the steps of acquiring an image including the precision approach channel indicator (S810), performing image preprocessing to perform object recognition on the precision approach channel indicator based on the image (S820), performing RGB masking on the precision approach channel indicator with object recognition to extract red light regions (S830), and applying image binarization and image contouring to calculate and display the number of extracted red light regions (S840).

[0112] In the image preprocessing step (S820), an object detection algorithm can be used to identify objects in the precision approach path indicator and set them as regions of interest.

[0113] Furthermore, in the step of extracting the red light area (S830), the threshold of the RGB value can be adjusted for the region of interest, thereby extracting the red light area from the precision approach path indicator identified by the object.

[0114] Furthermore, in the step of calculating and displaying the number of extracted red light areas (S840), for the precision approach channel indicator for object recognition with adjusted RGB value thresholds, image binarization is applied to display the extracted red light areas as white and the remaining areas as black. For the precision approach channel indicator for object recognition with applied image binarization, image contours are applied to extract the outer contour lines of the areas displayed as white, and the number of extracted outer contour lines is counted to calculate the number of extracted red light areas.

[0115] Furthermore, all the operational processes of the processor 130 as described above can be included in the control method of the signal status discrimination system of the precision approach path indicator according to an embodiment of the present invention.

[0116] In addition, a computer-readable storage medium may be provided, which stores a program for executing the control method of the signal status determination system of the precision approach path indicator described above in a computer.

[0117] Figure 9 It is shown Figure 1 The diagram shows the specific structure of the signal status determination system for a precision approach path indicator.

[0118] Reference Figure 9 The signal status determination system 100 of the precision approach path indicator may include a display 110, a camera unit 120, a processor 130, and a storage unit 140.

[0119] The processor 130 controls the operation of the signal status determination system 100 of the precision approach path indicator as a whole.

[0120] Specifically, the processor 130 includes RAM 131, ROM 132, main CPU 133, graphics processing unit 134, first interface 135-1 to nth interface 135-n, and bus 136.

[0121] RAM131, ROM132, main CPU133, graphics processing unit134, first interface 135-1 to nth interface 135-n, etc. can be connected to each other via bus 136.

[0122] The first interface 135-1 to the nth interface 135-n are connected to the various constituent elements described above. One of the multiple interfaces can be a network interface that connects to an external device via a network.

[0123] The main CPU 133 accesses the storage unit 140 and starts up using the O / S stored in the storage unit 140. In addition, it performs various actions using the various programs, contents, data, etc. stored in the storage unit 140.

[0124] In particular, the main CPU 133 can perform image preprocessing for object recognition of the precision approach channel indicator based on the image, perform RGB masking on the precision approach channel indicator for object recognition to extract the red light area, and apply image binarization and image contouring to calculate the number of extracted red light areas and display them on the display.

[0125] ROM 132 stores the instruction set used to boot the system. When a power-on command is input to supply power, the main CPU 133 copies the O / S (Operation / Skill) data stored in storage unit 140 to RAM 131 according to the instructions stored in ROM 132, and executes the O / S to boot the system. After booting, the main CPU 133 copies various application programs stored in storage unit 140 to RAM 131 and executes the copied application programs to perform various actions.

[0126] The graphics processing unit 134 uses an arithmetic unit (not shown) and a rendering unit (not shown) to generate a screen containing various objects such as icons, images, and text. The arithmetic unit (not shown) calculates the coordinates, shape, size, color, and other attribute values ​​of each object to be displayed based on the received control commands. The rendering unit (not shown) generates a screen containing various layouts of objects based on the attribute values ​​calculated by the arithmetic unit (not shown).

[0127] In particular, the graphics processing unit 134 can display objects generated by the main CPU 133 through GUI (Graphical User Interface), icons, user interface screens, etc.

[0128] On the other hand, the operation of the processor 130 described above can be implemented by a program stored in the storage unit 140.

[0129] The storage unit 140 stores various data, including the O / S (Operating System) software module used to drive the signal status discrimination system 100 for the precision approach path indicator, and various multimedia content.

[0130] In particular, the storage unit 140 may include a software module for performing image preprocessing to identify objects in the precision approach channel indicator based on the image, extracting red light areas by performing RGB masking on the precision approach channel indicator with identified objects, applying image binarization and image contouring, calculating the number of extracted red light areas and displaying them on the display.

[0131] Figure 10 This is a diagram of a software module stored in a storage unit according to an embodiment of the present invention.

[0132] Reference Figure 10 The storage unit 140 can store programs such as the image preprocessing module 141, the RGB mask module 142, the image binarization module 143, and the image contour module 144.

[0133] On the other hand, the operation of the processor 130 described above can be implemented by a program stored in the storage unit 140. The detailed operation of the processor 130 using the program stored in the storage unit 140 will be described in detail below.

[0134] The image preprocessing module 141 can use an object detection algorithm to identify objects in the precision approach path indicator and set them as regions of interest in the image acquired by the camera unit 120.

[0135] In addition, the RGB mask module 142 can adjust the threshold of the RGB values ​​for the region of interest, thereby extracting the red light area from the precision approach path indicator for object recognition.

[0136] In addition, the image binarization module 143 can apply image binarization to the precision approach path indicator identified by the object with the threshold of the RGB values ​​adjusted, so that the extracted red light area is displayed as white and the remaining area is displayed as black.

[0137] In addition, the image contour module 134 can apply image contours to the precision approach path indicator that has been identified by image binarization, thereby extracting the outer contour line of the area displayed in white.

[0138] On the other hand, although Figure 10Not shown, but as described above, it may further include an automatic correction deep learning model generation module based on environmental information, or a mask parameter generation module based on the automatic correction deep learning model based on environmental information, or an automatic correction module based on the automatic correction deep learning model based on environmental information, etc.

[0139] On the other hand, a non-transitory computer-readable medium can be provided, which stores a program that sequentially executes the control method according to the invention.

[0140] Non-transitory readable media refers to media that stores data semi-permanently and can be read by a device, unlike media such as registers, caches, and memory which only store data temporarily. Specifically, the aforementioned applications or programs can be provided on non-transitory readable media such as CDs, DVDs, hard drives, Blu-ray discs, USB drives, memory cards, and ROMs.

[0141] Furthermore, the bus is not shown in the above block diagram of the signal status discrimination system for the precision approach channel indicator, but it may also include processors such as CPUs and microprocessors to perform image preprocessing for object recognition of the precision approach channel indicator based on the image, extracting the red light area by performing RGB masking on the precision approach channel indicator for the object recognition, applying image binarization and image contouring, calculating the number of extracted red light areas and displaying them on the display.

[0142] Furthermore, the preferred embodiments of the present invention are shown and illustrated in the above figures, but the present invention is not limited to the specific embodiments described above. Those skilled in the art to which this invention pertains can make various modifications without departing from the spirit of the invention as claimed in the claims, and such modifications should not be understood separately from the technical concept or prospect of the present invention.

Claims

1. A signal status determination system for a precision approach path indicator, comprising: monitor; The camera unit acquires images including those from the aforementioned precision approach path indicator; as well as The processor performs image preprocessing based on the image to identify objects in the precision approach channel indicator. It performs RGB masking on the precision approach channel indicator to extract red light areas, applies image binarization and image contouring, calculates the number of extracted red light areas, and displays them on the display.

2. The signal status determination system for a precision approach path indicator according to claim 1, wherein, The processor uses an object detection algorithm to identify objects in the precision approach path indicator and sets them as regions of interest.

3. The signal status determination system for a precision approach path indicator according to claim 2, wherein, The processor adjusts the threshold of the RGB values ​​for the region of interest and extracts the red light area from the precision approach path indicator identified by the object.

4. The signal status determination system for a precision approach path indicator according to claim 3, wherein, The processor applies image binarization to the precision approach path indicator identified by the object with the threshold of the RGB values ​​adjusted, and displays the extracted red light area as white and the remaining area as black.

5. The signal status determination system for a precision approach path indicator according to claim 4, wherein, The processor extracts the outer contour of the area displayed in white from the image contour of the object identified by the above-mentioned application image binarization.

6. The signal status determination system for a precision approach path indicator according to claim 5, wherein, The processor counts the number of extracted outer contour lines to calculate the number of extracted red light areas.

7. The signal status determination system for a precision approach path indicator according to claim 6, wherein, The processor controls the display to show the signal status corresponding to the number of red light areas extracted above.

8. A control method for a signal status determination system of a precision approach path indicator, comprising: The steps for acquiring images including the aforementioned precision approach path indicator; Perform image preprocessing steps to identify objects in the precision approach path indicator based on the above images; The steps for extracting the red light area by performing RGB masking on the precision approach path indicator identified by the above-mentioned object; as well as The steps involve applying image binarization and image contouring to calculate and display the number of red light regions extracted above.

9. The control method for the signal status determination system of the precision approach path indicator according to claim 8, wherein, In the above image preprocessing steps, an object detection algorithm is used to identify objects in the precision approach path indicator and set them as regions of interest.

10. The control method for the signal status determination system of the precision approach path indicator according to claim 9, wherein, In the step of extracting the red light area, the threshold of the RGB value is adjusted for the region of interest, and the red light area is extracted from the precision approach path indicator identified by the object.

11. The control method for the signal status determination system of the precision approach path indicator according to claim 10, wherein, In the step of calculating and displaying the number of extracted red light areas, for the precision approach path indicator identified by the object with the threshold of the RGB values ​​adjusted, image binarization is applied to display the extracted red light areas as white and the remaining areas as black. For the image contour of the precision approach path indicator used in the above-mentioned object recognition using image binarization, the outer contour lines of the areas displayed as white are extracted, and the number of the extracted outer contour lines is counted to calculate the number of the extracted red light areas.

12. A computer-readable storage medium storing a program for executing in a computer a control method for a signal status determination system of a precision approach path indicator according to any one of claims 8 to 11.

Images