A computer video interface line electromagnetic radiation signal detection and image restoration method

Abstract

The present invention discloses a computer video interface line electromagnetic radiation signal detection and image restoration method, the steps are as follows: using wireless electromagnetic spectrum signal receiving equipment, combined with VESA video standards, for TEMPEST frequency points, according to the set sampling rate, the HDMI / VGA interface cable radiation signal is sampled; time-frequency conversion is performed on the sampled data, and parameters such as line period and field start position are extracted; on the premise of determining the line frequency and field start position, video period synchronization is performed to obtain field frequency parameters; line tracking search is performed to extract image frame data; grayscale processing and multi-frame overlapping processing are performed on the image frame data to restore the video image. The present invention simplifies the computer video interface line electromagnetic radiation signal detection method, quickly extracts line frequency parameters, field start position, etc., performs line tracking to extract image frame data, and meets the needs of restoring video images from the computer video interface line electromagnetic radiation signal.

Description

Technical Field

[0001] The invention relates to the technical field of computer electromagnetic leakage detection and restoration, and in particular to a computer video interface line electromagnetic radiation signal detection and image restoration method. Background Art

[0002] With the rapid development of electronic products and information technology and product iteration, computer equipment has become an indispensable part of daily life and work. As an indispensable tool in people's work, the information security of computer equipment is very important, especially the information security of computers for confidential work. During the operation of the computer, it needs to be connected to the monitor through a video interface cable for visual desktop display, and the video information will be hidden in the electromagnetic radiation.

[0003] Existing detection methods are unable to automatically and effectively detect radiation signals (including main frequency points and multiple harmonics) according to the usage scenarios of various computer devices, automatically select the best radiation signal, extract line frequency parameters, field start position, perform line tracking, extract image frame data, and restore the image. Summary of the invention

[0004] The invention provides a computer video interface line electromagnetic radiation signal detection and image restoration method, which can realize effective detection of electromagnetic leakage signals and effective restoration of video image information.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] A computer video interface line electromagnetic radiation signal detection and image restoration method, comprising the following steps:

[0007] Step S1: sampling the radiation signal of the HDMI / VGA interface cable at the TEMPEST frequency point according to the set sampling rate;

[0008] Step S2: Perform time-frequency conversion on the sampled data to confirm whether there is a valid signal. If so, proceed to step S3, otherwise return to step S1;

[0009] Step S3: perform a line period search on the sampled data, and search for line frequency parameters through sliding correlation. If the line frequency parameters are effectively extracted, execute step S4, otherwise return to step S1;

[0010] Step S4: Perform field synchronization processing on the sampled data. If the field start position is effectively extracted, execute step S5; otherwise, return to step S1;

[0011] Step S5: Perform video cycle synchronization on the sampled data. If the video cycle synchronization is successful, the field frequency parameters are obtained and step S6 is executed. Otherwise, the process returns to step S1.

[0012] Step S6: perform line tracking search on the sampled data, extract image frame data, and execute step S7;

[0013] Step S7: grayscale processing and multi-frame overlapping processing are performed on the image frame data to restore the video image.

[0014] Among them, step S1, refer to the frequency table of each pixel point defined in the DMT table of the VESA standard, for each TEMPEST frequency point and Subharmonic frequency, time domain sampling of the frequency band of the radiated signal of the HDMI / VGA interface cable.

[0015] Step S2: Perform fast Fourier transform on the sampled data to obtain the frequency domain information of the signal and extract each TEMPEST frequency point. The characteristic parameters of the radiated signal at , Amplitude , signal-to-noise ratio ,Will Reaching the threshold The frequency points form a frequency point set ,according to Values ​​in reverse order, if If it is not 0, go to step S3; otherwise, return to step S1.

[0016] Step S3, take The maximum frequency, that is, the frequency The sampling rate is 250MHz, and the sampling data is bandpass filtered. Taking into account the data transmission and restoration performance, 10 times sampling is performed to obtain 25MHz sampling data. The similarity between rows of image data is used to obtain the row period length using the sliding correlation method.

[0017] The details are as follows: each time the sampling rate is 25MHz Data of points, based on the data The points are the correlation coefficients, After the data is calculated by sliding correlation, the offset length corresponding to the position with the largest correlation value is the row period length. ,

[0018] ,

[0019] ,

[0020] in, is the sliding related value, is the sampling data amplitude, The maximum position is the line period length ,

[0021] In order to reduce the error, three consecutive Point data, repeat the sliding correlation calculation steps, if the three obtained If the error is within 20, the average of three times is used. is the final row cycle length.

[0022] Step S4, extracting the field start position, synchronizing the start chip position of the TMDS video frame, and taking the start position of the data island of the first line of the TMDS video frame as the field synchronization position,

[0023] The details are as follows:

[0024] (1) Data grouping: using 4 times the line period length obtained in step S3 as the chip group length, the time domain sampling data is grouped.

[0025] ,

[0026] in, For different grouping sequences, It is the subscript within the chip group value, and its value range is: ， For each group number, is the group length,

[0027] (2) Calculate the average value of the data amplitude of each group and the difference between the average values.

[0028] ,

[0029] in is the average value of each group, is the difference between the mean values ​​of each group,

[0030] (3) Search for data island intervals,

[0031] Use the difference between the means of each group To determine the location of the data island, the data between two data islands is a "field".

[0032] Pick The minimum interval of data island is 200 groups. Medium or larger The value of can be used to find the group number of each data island ,

[0033] (4) Calculate the group offset value where the data island is located ,

[0034] According to the obtained data island group number, take the average value of all groups between two data islands ,

[0035] , find the position where the group mean value drops significantly, which is the group offset value ,

[0036] (5) Calculate the offset value within the group at the start position of the field ,

[0037] Based on the group offset value obtained in step (4), take 4 groups of chip data on the left and right sides, perform smoothing filtering, obtain the time domain waveform, and find the end position of the control period T3 as the starting position of the data island, that is, the field starting position. .

[0038] Step S5, based on the extracted row cycle length and field start position, calculate the mean of each row of data, and determine the actual number of video rows by statistically analyzing the row mean sequence and vertical blanking lines ,

[0039] The details are as follows:

[0040] (1) Take the signal amplitude. Since the starting position of the field is fixed at 9 lines, extract Row data;

[0041] (2) Calculate the mean of each row of data , statistical mean sequence, minus the minimum mean to get the target sequence, that is

[0042] ;

[0043] (3) 2 / 3 of the maximum value of the sequence is used as the decision threshold, that is,

[0044] ;

[0045] (4) Find the first value less than of Location is , The number of vertical blanking lines ,extract The first row number above the threshold , The number of vertical resolution lines ,

[0046] The number of chips in the field is calculated as:

[0047] ,

[0048] The field frequency is obtained as:

[0049] ,

[0050] in is the data sampling rate, field frequency It's 50 or 60.

[0051] Step S6 is specifically as follows:

[0052] (1) Based on the synchronization offset value and the number of vertical blanking lines, the first line after the vertical blanking line is taken as the starting position of the video data;

[0053] (2) Determine the tracking window size based on the vertical blanking line, search for the maximum rising edge in the large window, and correct the starting position of the video data;

[0054] (3) Extract each line of video data, determine the tracking window size based on the line length, use the small window to search for the maximum rising edge, correct the starting position of each line, and record the correction value;

[0055] (4) Align the starting position of each row. When the number of extracted rows is greater than the vertical resolution, the image can be formed. If the tracking correction value changes by more than 10 code chips, the tracking is considered to have failed.

[0056] Step S7 is specifically as follows:

[0057] (1) Calculate the average value of all rows to obtain the average difference sequence, and use 2 / 3 of the maximum value as the threshold value.

[0058] (2) Find the first position that crosses the threshold As the video image start line, calculate the activation image end line ,

[0059] (3) Extract the middle row data, define the starting position of the activated image column resolution as the maximum falling edge trough position, and directly calculate the ending position of the activated image column resolution; according to the starting and ending positions of the resolution row and resolution column found, cut out the activated image, and for the blurred area, use multiple frames of image overlapping to synthesize the picture, which has the beneficial effect of improving the image clarity.

[0060] Compared with the prior art, the advantages of the present invention are as follows:

[0061] 1. The present invention can effectively detect electromagnetic leakage signals of computers.

[0062] 2. The present invention extracts line frequency and field frequency parameters from time domain data.

[0063] 3. The present invention tracks the leakage signal and effectively extracts the image frame data.

[0064] 4. The present invention performs grayscale processing on the image frame data and adopts multi-frame overlapping processing to restore the image clearly.

[0065] 5. The present invention provides a computer video interface line electromagnetic radiation signal detection and image restoration method, which can achieve effective detection of electromagnetic leakage signals and effective restoration of video image information. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is the flow chart of electromagnetic radiation signal detection and image restoration of the video interface line.

[0067] Figure 2 This is the electromagnetic signal diagram radiated by the video interface line.

[0068] Figure 3 is the line frequency characteristic diagram of the radiated electromagnetic signal,

[0069] Figure 4 is the group mean waveform of the radiated electromagnetic signal,

[0070] Figure 5 is the waveform diagram of the intra-group value of the radiated electromagnetic signal.

[0071] Figure 6 is the frequency characteristic diagram of the radiated electromagnetic signal.

[0072] Figure 7 It is the row peak characteristic diagram of the radiated electromagnetic signal. DETAILED DESCRIPTION

[0073] In order to deepen the understanding of the present invention, the present embodiment is described in detail below with reference to the accompanying drawings.

[0074] Example 1: See Figure 1 The present invention provides a computer video interface line electromagnetic radiation signal detection and image restoration method, comprising the following steps:

[0075] Step S1: sampling the radiation signal of the HDMI / VGA interface cable at the TEMPEST frequency point according to the set sampling rate;

[0076] Step S2: Perform time-frequency conversion on the sampled data to confirm whether there is a valid signal. If so, proceed to step S3, otherwise return to step S1;

[0077] Step S3: perform a line period search on the sampled data, and search for line frequency parameters through sliding correlation. If the line frequency parameters are effectively extracted, execute step S4, otherwise return to step S1;

[0078] Step S4: Perform field synchronization processing on the sampled data. If the field start position is effectively extracted, execute step S5; otherwise, return to step S1;

[0079] Step S5: Perform video cycle synchronization on the sampled data. If the video cycle synchronization is successful, the field frequency parameters are obtained and step S6 is executed. Otherwise, the process returns to step S1.

[0080] Step S6: perform line tracking search on the sampled data, extract image frame data, and execute step S7;

[0081] Step S7: grayscale processing and multi-frame overlapping processing are performed on the image frame data to restore the video image.

[0082] Step S1: referring to the frequency table of each pixel point defined in the DMT table of the VESA standard, for each TEMPEST frequency point and Subharmonic frequency (considering signal energy attenuation and propagation characteristics, the upper limit of harmonic frequency The value is 8) to perform time domain sampling on the frequency band of the radiated signal of the HDMI / VGA interface cable.

[0083] Step S2: Perform fast Fourier transform on the sampled data to obtain the frequency domain information of the signal and extract each TEMPEST frequency point The characteristic parameters of the radiated signal at , Amplitude , signal-to-noise ratio ,Will Reaching the threshold The frequency points form a frequency point set ,according to Values ​​in reverse order, if If it is not 0, go to step S3; otherwise, return to step S1.

[0084] Step S3: Take The maximum frequency, that is, the frequency The center frequency is used, and the time domain sampling is performed at a sampling rate of 250MHz. The sampled data is bandpass filtered, taking into account both data transmission and restoration performance, and 10 times sampling is performed to obtain 25MHz sampling data. The similarity between rows of image data is used to obtain the row period length using the sliding correlation method.

[0085] The details are as follows: each time the sampling rate is 25MHz Data of points, based on the data The points are the correlation coefficients, After the data is calculated by sliding correlation, the offset length corresponding to the position with the largest correlation value is the row period length. ,

[0086] ,

[0087] ,

[0088] in, is the sliding related value, is the sampling data amplitude, The maximum position is the line period length ,

[0089] In order to reduce the error, three consecutive Point data, repeat the sliding correlation calculation steps, if the three obtained If the error is within 20, the average of three times is used. is the final row cycle length.

[0090] Step S4: extracting the field start position and synchronizing the start chip position of the TMDS video frame. In this step, the start position of the data island of the first line of the TMDS video frame is used as the field synchronization position because the data amplitude corresponding to this position is the strongest, which is conducive to search and tracking.

[0091] Specifically,

[0092] (1) Data grouping: using 4 times the line period length obtained in step S3 as the chip group length, the time domain sampling data is grouped.

[0093] ,

[0094] in, For different grouping sequences, It is the subscript within the chip group value, and its value range is: ， For each group number, is the group length,

[0095] (2) Calculate the average value of the data amplitude of each group and the difference between the average values.

[0096] ,

[0097] in is the average value of each group, is the difference between the mean values ​​of each group,

[0098] (3) Search for data island intervals,

[0099] Use the difference between the means of each group To determine the location of the data island, the data between two data islands is a "field".

[0100] Pick The minimum interval of data island is 200 groups. Medium or larger The value of can be used to find the group number of each data island ,

[0101] (4) Calculate the group offset value where the data island is located ,

[0102] According to the obtained data island group number, take the average value of all groups between two data islands ,

[0103] , find the position where the group mean value drops significantly, which is the group offset value ,

[0104] (5) Calculate the offset value within the group at the start position of the field ,

[0105] It is known that in addition to the "video period T1", there are also "protection interval T2", "control period T3" and "data island period T4" in the radiation signal of the video interface cable, and the radiation signals in these three places are all in square wave form. Therefore, the starting position of the field can be determined by searching for the starting position of the data island square wave. Based on the group offset value obtained in step (4), 4 groups of code chip data are taken on the left and right, and smoothing filtering is performed to obtain the time domain waveform. The end position of the control period T3 is found as the starting position of the data island, that is, the starting position of the field. .

[0106] Step S5, based on the extracted row cycle length and field start position, calculate the mean of each row of data, and determine the actual number of video rows by statistically analyzing the row mean sequence and vertical blanking lines ,

[0107] The details are as follows:

[0108] (1) Take the signal amplitude. Since the starting position of the field is fixed at 9 lines, extract Row data;

[0109] (2) Calculate the mean of each row of data , statistical mean sequence, minus the minimum mean to get the target sequence, that is

[0110] ;

[0111] (3) 2 / 3 of the maximum value of the sequence is used as the decision threshold, that is,

[0112] ;

[0113] (4) Find the first value less than of Location is , The number of vertical blanking lines ,extract The first row number above the threshold , The number of vertical resolution lines ,

[0114] The fixed offset of 9 lines in step (1) is because each field data is basically 4 data island periods, followed by 5 control periods, so the known 9 line periods are offset first.

[0115] The number of chips in the field is calculated as:

[0116] ,

[0117] The field frequency is obtained as:

[0118] ,

[0119] in is the data sampling rate, field frequency Usually it is 50 or 60, and the reason for rounding here is is an instantaneous integer value, the corresponding real value may be a decimal, there is a certain error, and the real field frequency It is an integer, usually 50 or 60.

[0120] Step S6: The amplitude of the peak position at the start of the guard interval is the strongest, and the corresponding rising edge amplitude changes the most. The end position of the maximum rising edge in the search window is the tracking position. The tracking position must be aligned each time the search is performed. To ensure the reliability of the system, the tracking result is judged. If the tracking correction value changes by more than 10 chips, it is considered that the tracking has failed. Then the field synchronization and video cycle synchronization processes are re-executed to find the starting position and length value of the video data, and the video data is re-extracted.

[0121] Specifically,

[0122] (1) Based on the synchronization offset value and the number of vertical blanking lines, the first line after the vertical blanking line is the starting position of the video data;

[0123] (2) Determine the tracking window size based on the vertical blanking line, search for the maximum rising edge with a large window, and correct the starting position of the video data;

[0124] (3) Extract each line of video data, determine the tracking window size based on the line length, use the small window to search for the maximum rising edge, correct the starting position of each line, and record the correction value;

[0125] (4) Align the starting position of each row. When the number of extracted rows is greater than the vertical resolution, the image can be formed. If the tracking correction value changes by more than 10 code chips, the tracking is considered to have failed.

[0126] The step S7: after completing the row-field synchronization, video cycle synchronization and row tracking search, a two-dimensional image matrix is ​​obtained, and the pixel positions between rows are aligned. According to the sending situation of the "activation image", when the parameters are unknown, the starting position needs to be calculated to extract the activation image. Only the starting position of the first frame image needs to be calculated, and the subsequent images can be directly extracted. For various resolution rows, not all rows carry data, and there are certain blank images at the beginning and end. Since the radiation value when switching from the "blank image row" to the "video image row" is the largest, the starting row of the video image can be obtained by row average calculation; for various resolution columns, the field synchronization and tracking positions are the starting positions of the "guard interval", and the resolution row needs to offset the length of the guard interval. After synchronous search and tracking, a complete "TMDS" video frame is extracted, that is, the image data of a field. Different RGB values ​​of the original HDMI / VGA image correspond to different signal amplitudes and different radiation energy values. The grayscale image of the original image is restored by the amplitude value of the signal collected by the air interface. For the partially blurred areas in the image, the multi-frame image overlap method is used to improve the image clarity.

[0127] Specifically,

[0128] (1) Calculate the average value of all rows to obtain the average difference sequence, and use 2 / 3 of the maximum value as the threshold value.

[0129] (2) Find the first position that crosses the threshold As the starting line of the video image, since the blank image is symmetrical, calculate the end line of the active image ,

[0130] (3) Extract the middle row data. Since the radiation intensity of the guard interval is slightly larger, the starting position of the activated image column resolution is defined as the maximum falling edge trough position. Due to the symmetric characteristics of the guard interval, the end position of the activated image column resolution can be directly calculated.

[0131] (4) According to the start and end positions of the resolution rows and resolution columns found, the activation image is cut out. Since the row-to-column ratio of the extracted activation image generally does not conform to the normal display ratio, it is also necessary to downsample the data for averaging to reduce image noise. At different resolutions, the row-to-column ratio is close to 16:9, and at the 25Mhz sampling rate used in this application, the row length of the activation image obtained is smaller than the column length, so only each column of data is downsampled here, and the row data is not downsampled;

[0132] Since adjacent image frames are similar, for blurred areas, multiple frames of images can be overlapped to synthesize the image and improve image clarity.

[0133] Embodiment 2: A method for detecting electromagnetic radiation signals and restoring images of a computer video interface line comprises the following steps: The flowchart of the scheme is as follows: Figure 1 As shown,

[0134] The detection range of the wireless electromagnetic spectrum receiver is set to 20MHz~1500MHz, and the data acquisition sampling rate is The computer video interface cable uses an HDMI cable; the display resolution is 1920*1080 pixels, and the refresh rate is 60Hz.

[0135] Step S1: Refer to the frequency table of each pixel point defined in the DMT table of the VESA standard and detect the TEMPEST frequency points respectively and Subharmonic frequency (considering signal energy attenuation and propagation characteristics, the upper limit of harmonic frequency The value is 8) to perform time domain sampling on the frequency band of the radiated signal of the HDMI / VGA interface cable.

[0136] Step S2: By setting different AD sampling center frequencies, using a 250M sampling rate, sampling cable radiation signals, performing time-frequency conversion on the sampled data, and extracting each TEMPEST frequency point The bandwidth of the radiated signal , Amplitude , signal-to-noise ratio ,according to Arrange in reverse order to form a frequency point set .

[0137] Step S3: Frequency The sampling data is band-pass filtered and decimated by 10 times to obtain 25MHz sampling data. The data spectrum is shown in the attached figure. Figure 2 As shown, the sliding correlation line period search method is used to continuously perform sliding correlation on three frames of data, and the average peak interval is 372, that is, the image data line length at a sampling rate of 25MHz is obtained. The sliding correlation value of the first frame data is shown in the attached figure. Figure 3 As shown, the horizontal coordinates of the two points that pass the threshold are 1615 and 1989, so the row length of this frame data is 1989-1615=374. Similarly, the row lengths of the last two frames of data are 371 and 371 respectively. The average of the three values ​​is (374+371+371) / 3=372.

[0138] Step S4: Extract the starting chip position of the field, take 4 times the line length as the group length, calculate the average of each group, perform statistical analysis on the mean sequence, and take the first group greater than the threshold as the group offset value The group offset value is 89, and the mean sequence waveform is as shown in the attached figure. Figure 4 As shown in the figure, the obvious rising edge is at position 89. The part with the highest amplitude in the waveform is the square wave of the "data island period", and the square wave immediately before the "data island period" is the "control period". In the four groups of waveforms before and after group 89, the position of the first code chip greater than the threshold is found as the intra-group offset. The waveform diagram of the data before and after group 89 is shown in the attached figure. Figure 5 As shown, the first chip position greater than the threshold is 5400, which is converted to an intra-group offset value of 5400-4*4*372 = -552. Therefore, the field start position is 88*4*372-552 = 130392.

[0139] Step S5: Based on the extracted line length and field start position, the actual number of video data lines is obtained by statistically analyzing the line mean sequence. 994, vertical blanking lines is 134, and the number of chips in the field is calculated = (994+134) * 372 = 419616, field frequency Attached Figure 6 It is a group mean sequence waveform, showing a periodic pattern. The square wave area with higher amplitude is the vertical blanking part.

[0140] Step S6: Perform row tracking search on the data. The end position of the maximum rising edge in the search window is the tracking position. If the tracking correction value changes by less than 10 chips, the tracking is successful. Figure 7 It is the amplitude waveform of each row. There is an obvious peak at the same position of each row. The tracking position can be searched based on the maximum rising edge.

[0141] Step S7: Extract the "TMDS" video frame (a two-dimensional image matrix), that is, the image data of a field. According to the different resolutions of the original image and the different characteristics of the air interface radiation code rate corresponding to the refresh rate, a 250MHz sampling rate is used to meet the different resolutions. The extracted video data is downsampled and averaged to reduce image noise. The default smoothing multiples are R=8 and C=3. For brightness processing, the default "pixel mean-0.5 grayscale" is used to adaptively restore the image information based on the grayscale value of the current image to reduce the impact of the original image color on the grayscale value. Considering that there are some blurred areas in the image, 8 frames of image overlap are used to improve the local clarity of the image.

[0142] It should be noted that the above embodiments are not intended to limit the protection scope of the present invention, and equivalent changes or substitutions made on the basis of the above technical solutions all fall within the protection scope of the claims of the present invention.

Claims

1. A computer video interface line electromagnetic radiation signal detection and image restoration method, characterized in that: The steps include: Step S1: sampling the radiation signal of the HDMI / VGA interface cable at the TEMPEST frequency point according to the set sampling rate; Step S2: Perform time-frequency conversion on the sampled data to confirm whether there is a valid signal. If so, proceed to step S3, otherwise return to step S1; Step S3: perform a line period search on the sampled data, and search for line frequency parameters through sliding correlation. If the line frequency parameters are effectively extracted, execute step S4, otherwise return to step S1; Step S4: Perform field synchronization processing on the sampled data. If the field start position is effectively extracted, execute step S5; otherwise, return to step S1; Step S5: Perform video cycle synchronization on the sampled data. If the video cycle synchronization is successful, the field frequency parameters are obtained and step S6 is executed. Otherwise, the process returns to step S1. Step S6: perform line tracking search on the sampled data, extract image frame data, and execute step S7; Step S7: grayscale processing and multi-frame overlapping processing are performed on the image frame data to restore the video image.

2. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S1, referring to the frequency table of each pixel point defined in the DMT table of the VESA standard, for each TEMPEST frequency point and Subharmonic frequency, time domain sampling of the frequency band of the radiated signal of the HDMI / VGA interface cable.

3. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S2: Perform fast Fourier transform on the sampled data to obtain the frequency domain information of the signal and extract each TEMPEST frequency point. The characteristic parameters of the radiated signal at , Amplitude , signal-to-noise ratio ,Will Reaching the threshold The frequency points form a frequency point set ,according to Values ​​in reverse order, if If it is not 0, go to step S3; otherwise, return to step S1.

4. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S3, take The maximum frequency, that is, the frequency As the center frequency, the time domain sampling is performed at a sampling rate of 250MHz, the sampled data is bandpass filtered, and 10 times decimation is performed to obtain 25MHz sampled data. The sliding method is used to obtain the line period length. The details are as follows: each time the sampling rate is 25MHz Data of points, based on the data The points are the correlation coefficients, After the data is calculated by sliding correlation, the offset length corresponding to the position with the largest correlation value is the row period length. , ， ， in, For each group number, is the subscript within the chip group value, is the sliding related value, is the sampling data amplitude, The maximum position is the line period length , take three times in a row Point data, repeat the sliding correlation calculation steps, if the three obtained If the error is within 20, the average of three times is used. is the final row cycle length.

5. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S4, extracting the field start position, synchronizing the start chip position of the TMDS video frame, and taking the start position of the data island of the first line of the TMDS video frame as the field synchronization position, The details are as follows: (1) Data grouping: using 4 times the line period length obtained in step S3 as the chip group length, the time domain sampling data is grouped. ， in, For different grouping sequences, It is the subscript within the chip group value, and its value range is: ， For each group number, is the group length, (2) Calculate the average value of the data amplitude of each group and the difference between the average values. ， in is the average value of each group, is the difference between the mean values ​​of each group, (3) Search for data island intervals, Use the difference between the means of each group To determine the location of the data island, the data between two data islands is a field. Pick The minimum interval of data island is 200 groups. Medium to large The value of can be used to find the group number of each data island , (4) Calculate the group offset value where the data island is located , According to the obtained data island group number, take the average value of each group between two data islands , , find the position where the group mean value drops significantly, which is the group offset value , (5) Calculate the internal offset value of the field start position , Based on the group offset value obtained in step (4), take 4 groups of chip data on the left and right sides, perform smoothing filtering, obtain the time domain waveform, and find the end position of the control period T3 as the starting position of the data island, that is, the field starting position. .

6. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S5, based on the extracted row cycle length and field start position, calculate the mean of each row of data, and determine the actual number of video rows by statistically analyzing the row mean sequence and vertical blanking lines , The details are as follows: (1) Take the signal amplitude, offset 9 lines from the start position of the field, and extract Row data; (2) Calculate the mean of each row of data , statistical mean sequence, minus the minimum mean to get the target sequence, that is ； (3) 2 / 3 of the maximum value of the sequence is used as the decision threshold, that is, ； (4) Find the first value less than of Location is , The number of vertical blanking lines ,extract The first row number above the threshold , The number of vertical resolution lines , The number of chips in the field is calculated as: ， The field frequency is obtained as: ， in is the data sampling rate, field frequency It's 50 or 60.

7. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S6 is specifically as follows: (1) Based on the synchronization offset value and the number of vertical blanking lines, the first line after the vertical blanking line is taken as the starting position of the video data; (2) Determine the tracking window size based on the vertical blanking line, search for the maximum rising edge in the large window, and correct the starting position of the video data; (3) Extract each line of video data, determine the tracking window size based on the line length, use the small window to search for the maximum rising edge, correct the starting position of each line, and record the correction value; (4) Align the starting position of each row. When the number of extracted rows is greater than the vertical resolution, the image can be formed. If the tracking correction value changes by more than 10 code chips, the tracking is considered to have failed.

8. A computer video interface line electromagnetic radiation signal detection and image restoration method according to claim 1, characterized in that: Step S7 is specifically as follows: Calculate the average value of all rows to obtain the average difference sequence, and use 2 / 3 of the maximum value as the threshold value; Find the first position that crosses the threshold As the video image start line, calculate the activation image end line , Extract the middle row data, define the starting position of the activated image column resolution as the maximum falling edge trough position, and directly calculate the ending position of the activated image column resolution; According to the start and end positions of the resolution row and resolution column found, the active image is cut out. For the blurred area, multiple frames of images are overlapped to synthesize the image to improve the image clarity.

Images