A rapid imaging method for marine 2D seismic data

Abstract

This invention discloses a rapid imaging method for 2D seismic data at sea, relating to the field of seismic exploration data processing technology. The method includes: acquiring raw SEGD format seismic data of the target area from a seismic acquisition system; merging shot data from each survey line and decompiling it into usable internal data to obtain a data volume to be processed; and loading the observation system based on the decompiled data volume and assigning seismic trace parameters according to the actual acquisition parameters. This invention effectively suppresses various types of marine interference through a multi-level denoising and signal protection strategy. It employs frequency-division multi-channel statistical anomaly amplitude suppression technology, combined with zero-phase high-pass filtering, to remove low-frequency strong-energy background noise while protecting effective low-frequency signals. A direct-wave cutoff step further eliminates strong-energy interference under long cable conditions, highlighting the effective reflection signals of mid-to-long-range offset data, forming a progressive signal-to-noise ratio improvement mechanism from raw data to final imaging, thus improving the interpretability of seismic data.

Description

Technical Field

[0001] This invention relates to the field of seismic exploration data processing technology, specifically a rapid imaging method for two-dimensional seismic data at sea. Background Technology

[0002] In the fields of marine geological surveys and offshore engineering, marine 2D multichannel seismic acquisition technology is an important means of exploring seafloor stratigraphic structure, identifying fault structures, discovering buried paleochannels, and assessing geological hazards such as shallow gas, seafloor collapse, and landslides. This technology is widely used in marine engineering fields such as waterway construction and pipeline inspection, as well as resource surveys such as marine sand resource surveys and natural gas hydrate exploration. However, marine 2D seismic data acquisition operations face many challenges: complex operation techniques, numerous environmental interference factors, high uncertainty, and high costs. To ensure construction quality and optimize acquisition parameters, it is necessary to perform rapid imaging processing on the real-time acquired seismic data.

[0003] Rapid imaging technology for marine 2D seismic data has become an indispensable part of marine operations due to its high efficiency and intuitiveness. This technology not only provides a basis for optimizing and adjusting acquisition parameters but also helps researchers quickly establish a preliminary understanding of seabed strata, structures, and sedimentary phenomena. However, the seismic data acquisition process is susceptible to various interference factors: on the one hand, external interference such as waves, noise from neighboring vessels, ship machinery vibration, seabirds carrying foreign objects, and cable leakage are difficult to completely avoid; on the other hand, natural factors such as water depth variations and rugged seabed topography in the survey area also affect data quality. These interference factors significantly reduce the signal-to-noise ratio and resolution of seismic data, thus affecting the accurate identification of seabed strata structures. Therefore, given the relative scarcity of rapid 2D multichannel seismic imaging methods for marine operations, this paper proposes a rapid imaging method for marine 2D seismic data, providing a new approach and solution to address the aforementioned technical challenges. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a rapid imaging method for two-dimensional seismic data at sea, comprising the following steps:

[0005] Step 1: Obtain raw SEGD format seismic data of the target area from the seismic acquisition system, merge the shot data of each survey line and decompile them into usable internal data for subsequent processing, and obtain the data body to be processed;

[0006] Step 2: Based on the decompiled data body to be processed, load the observation system and assign seismic trace head parameters according to the actual acquisition parameters, including trace number, shot distance, trace spacing, sampling rate, minimum offset, record length and other parameters, to establish an accurate spatial and temporal correspondence and guide subsequent processing;

[0007] Step 3: Based on the loading of the observation system, perform frequency division multi-channel statistical anomaly amplitude suppression and high-pass filtering to remove low-frequency strong energy interference, while protecting the effective low-frequency signal and improving the signal-to-noise ratio;

[0008] Step 4: In response to strong interference from direct waves in shallow water conditions of long cables, accurately cut off the mid-to-long offset direct waves in the filtered data to eliminate interference with ground reflection waves, highlight the effective signal, and ensure that the reflected signal is clear in order to obtain the data after the direct waves are cut off.

[0009] Step 5: Based on the data after direct wave shearing, a precise superimposed velocity field is established using high-order velocity analysis and multiple technical means. Through correction, compatible imaging of reflection interfaces with different tilt angles is achieved, and the superimposed effect is optimized.

[0010] Step 6: Based on the velocity analysis, perform dynamic correction and initial stacking, and implement amplitude energy compensation on the stacked profile to enhance the deep reflection signal. Analyze the amplitude attenuation characteristics, select the optimal parameters to further compensate for the deep reflection energy, enhance the continuity of the same axis, and finally output high-resolution two-dimensional seismic imaging results.

[0011] Preferably, step 1 specifically includes:

[0012] The raw seismic data of the target area is obtained from the seismic acquisition system. The raw seismic data is in SEGD format, and each survey line contains multi-shot data to ensure that the raw data format is uniform and the acquired information is complete, laying the foundation for subsequent processing.

[0013] The SEGD format multi-shot data corresponding to each survey line are merged to generate a continuous data file for a single survey line, realizing the spatiotemporal sequential integration of multi-shot data and improving data continuity and processability;

[0014] The merged continuous data files are decompiled and converted into an internally usable data format. Initial verification of the track header information is performed to form a data body to be processed. This data is then converted into a system-recognizable format and key parameters are verified to improve data quality and the reliability of subsequent processing.

[0015] Preferably, step 1 further includes:

[0016] The merging process is carried out on a survey line basis, reading SEGD data shot by shot and splicing them according to the shot sequence, so as to realize the continuous splicing of data according to the time-space sequence and form a single survey line data file.

[0017] The decompilation process parses the SEGD header and data body of continuous data files, converts them into a format that can be recognized by the internal software, and unifies the original data format into structured data that can be directly processed by the software.

[0018] The decompiled data undergoes a trace head information integrity check and is compared with the collected parameter records to verify the accuracy of the collected information and confirm that the data is accurate and usable. The collected parameters include at least the shot point coordinates, receiver point coordinates, sampling interval, and recording length to ensure that the data is accurate and reliable in both spatial and temporal attributes, laying a reliable foundation for subsequent processing.

[0019] Preferably, step 2 specifically includes:

[0020] Based on the data volume to be processed formed after decompilation, the observation system is loaded according to the actual acquisition scheme to realize the standardized mapping of the spatial geometry and time series of the acquired data, providing a unified basic framework for subsequent steps;

[0021] The system assigns values ​​to the trace head information of the seismic data. The key parameters for this assignment include the number of traces, shot distance, trace spacing, sampling rate, minimum offset, and record length. This ensures the consistency of key acquisition parameters and avoids distortion in subsequent processing due to incorrect or missing parameters.

[0022] By assigning values ​​to establish a precise correspondence between the spatial location and recording time of seismic traces, an observation system loading profile is generated, providing a spatial reference for subsequent processing and forming an intuitive spatiotemporal distribution diagram. This helps to determine whether the data arrangement is reasonable and facilitates subsequent filtering, cropping, and other processing.

[0023] Preferably, step 3 specifically includes:

[0024] High-pass filtering is applied to the seismic data after loading the observation system to suppress high-frequency noise while protecting the effective low-frequency signal components according to a pre-set cutoff threshold, significantly reducing high-frequency environmental noise interference and effectively preserving low-frequency geological reflection characteristics.

[0025] To address the spatiotemporal consistency of low-frequency, high-energy anomalous amplitude interference caused by background noise such as swells, a frequency division processing strategy is adopted to effectively suppress low-frequency, high-energy noise and improve the spatiotemporal consistency of the data.

[0026] By setting time windows and amplitude thresholds in different frequency bands, multichannel statistical methods are applied to identify and suppress abnormal amplitudes, effectively suppressing both regular and irregular strong noise and protecting low-frequency signals. Abnormal amplitudes are precisely suppressed, and low-frequency effective reflections are fully protected, thereby enhancing the overall quality of the data.

[0027] Preferably, step 4 specifically includes:

[0028] Identify and locate the direct wave that propagates directly from the earthquake source through the seawater layer to the seismic tow cable. The direct wave has the characteristics of short propagation path and high energy under shallow water and long cable acquisition conditions, which provides a clear interference target for subsequent accurate removal and avoids misjudgment of effective signals.

[0029] Based on the propagation distance curve of the direct wave, determine its distribution range in the seismic record, including its performance in the mid-to-far offset gathers, accurately define the cut-off area, and avoid unnecessary loss of shallow reflections.

[0030] Based on the aforementioned distribution range, the corresponding regions of the filtered data are precisely cut off to eliminate the interference and shielding of direct waves on the ground reflection signals, thereby significantly improving the signal-to-noise ratio and phase axis continuity of the effective reflected waves.

[0031] Preferably, step 4 further includes:

[0032] When determining the direct wave cut-off range, the changes in water depth and the ruggedness of the seabed topography in the survey area are taken into account. Based on the relationship between water depth, cable length and offset distance, the direct wave cut-off time window is dynamically calculated to realize the dynamic adaptive adjustment of the time window parameters and reduce the impact of topographic undulations on the cut-off effect.

[0033] For rugged seafloor topography, the cut parameters within the common center point (CDP) gather are adjusted to regulate the cut range, avoiding excessive loss of effective seafloor reflection signals due to improper cuts. This effectively protects the reflection information of undulating seafloor areas, improves the accuracy of stratigraphic imaging, and in field processing where efficiency is paramount, the cut range of far-offset seismic traces can be increased to thoroughly suppress direct wave interference, thereby obtaining data after direct wave cuts. This significantly suppresses the remnants of direct waves in far-offset areas and improves the visibility of deep reflections.

[0034] Preferably, step 5 specifically includes:

[0035] On the data after direct wave cut-off, velocity analysis points are selected, and high-order velocity analysis method is used to perform velocity analysis in order to reduce the impact of large offset velocity changes on deep velocity estimation, significantly improve the velocity picking accuracy of deep reflection signals, and reduce large offset time difference correction error.

[0036] By comprehensively utilizing velocity spectrum analysis, constant velocity scanning and variable velocity scanning techniques, and combining them with corresponding superimposed profiles and isovelocity profiles for cross-validation, the reliability and geological rationality of the velocity model are enhanced through multi-technology cross-validation.

[0037] Through the interactive verification and iterative optimization, an accurate superimposed velocity field model that can adapt to changes in longitudinal and lateral velocity is established, and a high-precision velocity field is constructed to provide reliable velocity support for subsequent dynamic correction and imaging.

[0038] Preferably, step 5 further includes:

[0039] By selecting key velocity analysis points on the velocity spectrum, dense sampling and fine analysis are performed on areas with weak deep reflection signals and low signal-to-noise ratios, which significantly improves the spatial resolution and stability of the deep velocity field and supports accurate imaging of effective deep reflections.

[0040] The common center point (CDP) gather data were dynamically corrected using the established stacking velocity field model. This dynamic correction process ensures that reflection interfaces with different dip angles are compatible when stacked in the same phase, laying the foundation for imaging complex geological structures and subsequent stacking processing. It effectively flattens the phase axis of reflection interfaces with different dip angles and significantly improves the signal-to-noise ratio and structural imaging quality of the stacked profile.

[0041] Preferably, step 6 specifically includes:

[0042] The dynamically corrected velocity field is applied to the initial superposition of the data to generate an initial superposition profile, which significantly improves the continuity of the reflection phase axis and the signal-to-noise ratio.

[0043] This study analyzes the amplitude attenuation characteristics of seismic waves during propagation caused by spherical diffusion, stratum absorption, interface transmission, and noise. It also focuses on the energy attenuation law of deep reflection, accurately quantifies the energy loss law, and provides a reliable basis for targeted compensation.

[0044] Based on the amplitude attenuation characteristic analysis results, the optimal amplitude compensation parameters are selected to perform amplitude energy compensation on the initial stacked profile in order to enhance the energy of deep reflection signals. Finally, a high-resolution two-dimensional seismic imaging profile is output, which effectively recovers the energy of weak deep signals and improves the overall profile imaging quality and resolution.

[0045] This invention provides a rapid imaging method for marine 2D seismic data. It offers the following advantages:

[0046] (I) This rapid imaging method for marine 2D seismic data effectively suppresses various marine interferences through a systematic multi-level denoising and signal protection strategy. It adopts frequency-division multi-channel statistical anomaly amplitude suppression technology, combined with zero-phase high-pass filtering, to remove low-frequency strong energy background noise while protecting low-frequency effective signals. The direct wave cut-off step further eliminates strong energy interference under long cable conditions, highlights the effective reflection signals of mid-to-long distance offset data, and forms a progressive signal-to-noise ratio improvement mechanism from raw data to final imaging, thereby improving the interpretability of seismic data.

[0047] (II) This rapid imaging method for marine 2D seismic data comprehensively utilizes high-order velocity analysis, multi-technology cross-validation, and iterative optimization strategies to establish an accurate stacked velocity field that adapts to changes in longitudinal and lateral velocities. For deep weak signal areas and structurally complex areas, it adopts methods such as densifying velocity analysis points, expanding coverage, and local fine scanning to effectively improve the lateral representativeness and deep accuracy of velocity extraction. Through interactive verification of velocity spectrum, constant velocity scanning, variable velocity scanning, and stacked profiles, it ensures the physical rationality and spatial consistency of the velocity model, laying a reliable velocity foundation for subsequent dynamic correction and stacked imaging.

[0048] (III) This rapid imaging method for marine two-dimensional seismic data implements systematic time-varying and space-varying energy compensation in the amplitude processing stage. Based on accurate amplitude attenuation characteristic analysis, dynamic gain coefficients are set for shallow, medium and deep layers respectively, and frequency domain shaping technology is introduced to avoid high-frequency noise amplification. This compensation strategy can effectively recover the deep reflection energy lost due to factors such as spherical diffusion and stratum absorption, significantly improve the continuity and recognizability of deep phase axes, and enhance the ability to depict geological information of the entire profile, especially the deep region. Attached Figure Description

[0049] Figure 1 This is a schematic diagram illustrating the workflow of a rapid imaging method for two-dimensional seismic data at sea according to the present invention.

[0050] Figure 2 This is a cross-sectional view of the observation system after the present invention is loaded (the key term for the track head is cdp, and the secondary key term is offset).

[0051] Figure 3 This is a cross-sectional view of a single shot before filtering in this invention;

[0052] Figure 4 This is a cross-sectional view of a single shot after filtering according to the present invention;

[0053] Figure 5 This is a single-channel cross-sectional view before filtering in this invention;

[0054] Figure 6 This is a single-channel cross-sectional view after filtering according to the present invention;

[0055] Figure 7 This is a cross-sectional view showing the difference in amplitude before and after the suppression of abnormal amplitude in this invention.

[0056] Figure 8 This is a cross-sectional view of the present invention before direct wave cut-off and dynamic correction;

[0057] Figure 9 This is a cross-sectional view of the direct wave cut-off and dynamic correction of the present invention;

[0058] Figure 10The velocity analysis diagram of the present invention is shown in (a) and (b) is the velocity picking diagram.

[0059] Figure 11 This is a superimposed cross-sectional view before amplitude energy compensation in this invention;

[0060] Figure 12 This is a superimposed cross-sectional view after amplitude energy compensation according to the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Example 1, please refer to Figure 1 This invention provides a technical solution: a rapid imaging method for two-dimensional seismic data at sea, comprising the following steps:

[0063] Step 1: Obtain raw SEGD format seismic data of the target area from the seismic acquisition system, merge the shot data of each survey line and decompile it into usable internal data for subsequent processing, and obtain the data body to be processed. In marine 2D seismic exploration operations, operators first obtain raw seismic data of the target area from the seismic acquisition system (SercelSeal428 or IONI / O system), storing it in the internationally recognized SEGD format (SEG-DRev3.0). Each survey line contains hundreds to thousands of shot data, and the number of channels recorded for each shot depends on the cable configuration. During the acquisition process, key parameters such as GPS time, shot coordinates, receiver location, sampling interval, and recording length (12030ms) for each shot are recorded in real time. After obtaining the raw seismic data, data merging and decompiling are performed on a survey line basis. In actual operation, technicians use the data input module of the seismic processing software to read the SEGD file one shot at a time in sequence according to the shot number and merge them to generate a continuous, time-space sequence-organized single survey line data file. During the merging process, the shot points are automatically verified. Consistency with the receiving point is checked, and anomalies such as missing or duplicate data are detected. Subsequently, the merged SEGD file is converted into a data format recognizable by the software through a decompilation program. At the same time, the trace head information is extracted and parsed, including shot point coordinates (using the WGS84 coordinate system), receiver coordinates, channel number, sampling rate (2ms), record length, and offset. At this stage, the parsed trace head information is compared with the acquisition log to verify its accuracy and completeness, forming a clearly structured data body ready for subsequent processing. After decompilation, the trace head information of each seismic trace is checked one by one to ensure that the key parameters (trace spacing 12.5m, shot spacing 37.5m, minimum offset 170m) are consistent with the actual acquisition scheme. At the same time, parameters such as coordinate system, time stamp, and channel gain are standardized to eliminate inconsistencies that may be caused by differences in acquisition equipment or configuration changes. This is done by combining automated scripts with manual verification to ensure that the data corresponds accurately in the spatial and temporal dimensions. The verified data body serves as the input basis for all subsequent processing steps.

[0064] Furthermore, step 1 also includes: In actual operation, using the survey line as the basic unit, the data input module of professional seismic processing software reads the original files conforming to the SEG-D Rev3.0 standard shot by shot according to the shot point sequence. The software automatically performs time-space stitching based on the timestamp and shot number sequence to generate a continuous and unified single-segment SEG-D format data stream. At the same time, it checks the consistency of the number of receiver points and sampling points recorded for each shot, and automatically marks any anomalies such as shot point duplication, data breakpoints, or temporal sequence errors, following seismic data processing specifications to ensure the spatiotemporal continuity and integrity of the merged data; the trace head segment and data volume of the merged data file are parsed separately: extracting and converting key trace head information including shot point coordinates (using the WGS84 geodetic coordinate system, with an accuracy better than 0.1 meters), receiver coordinates (relative positioning accuracy up to decimeter level), channel number, sampling interval (set to 2 milliseconds), recording length (usually 12030 milliseconds), and offset, etc. The data body is reassembled in floating-point format and stored in the system's internal database. During this process, the syntax and semantics of the track head segments are automatically checked to ensure the accuracy of the decompilation and conversion and the format compatibility, generating structured track gather data. The track head information extracted after decompilation is compared one by one with the actual acquisition parameter records to verify the rationality of the survey line distribution of the shot point and receiver point coordinates, whether the minimum offset distance meets the design requirements (controlled within 170 meters), and whether the sampling interval and recording length are consistent with the acquisition log. At the same time, the integrity and standardization of parameters such as coordinate system, time synchronization flag, and instrument gain are checked. The verification is carried out through a combination of automated scripts and manual interactive review, ultimately forming a standardized data body with accurate spatial-temporal attributes that can be directly used for velocity analysis, dynamic and static correction, and other processing procedures.

[0065] Step 2: Based on the decompiled data body to be processed, load the observation system and assign seismic trace head parameters according to the actual acquisition parameters, including trace number, shot distance, trace spacing, sampling rate, minimum offset, and recording length, to establish an accurate spatial-temporal correspondence and guide subsequent processing. Load the observation system and assign seismic trace head information according to the actual acquisition scheme. In actual operation, based on the acquisition parameter file, the trace number is set to 600, consistent with the cable configuration; the shot distance is fixed at 37.5 meters; the trace spacing is 12.5 meters; the sampling rate is set to 2 milliseconds; the minimum offset is 170 meters; and the recording length is uniformly 12030 milliseconds. Simultaneously, the trace head information in the data body to be processed is read and associated with the acquisition log to achieve automatic matching and filling of key parameters. After assignment, a corresponding observation system loading profile is generated, showing the spatial distribution relationship between shot points and receiver points. Through trace head assignment, a precise correspondence between the spatial location and recording time of seismic traces is further established, based on shot point coordinates (WGS84 coordinate system, with a plane accuracy better than...). The spatial offset of each seismic trace is calculated using the coordinates of the source and receiver points (0.1 meters) and the relative positioning accuracy down to the decimeter level. This offset is then correlated with the recorded time series. Simultaneously, based on the sampling rate (2 milliseconds) and the recording length (12030 milliseconds), the two-way travel time range corresponding to each time sampling point is determined, ensuring that each seismic trace has a unique and accurate spatial position and time marker on the profile. After the observation system is loaded, the assignment results are quality controlled to ensure the accuracy of the spatial-temporal correspondence. In actual operation, the integrity and rationality of the trace head parameters are checked through automated scripts to verify whether the shot distance is 37.5 meters and whether the minimum offset is strictly controlled within 170 meters. At the same time, the loading profile of the observation system is manually reviewed to confirm whether the distribution of shot points and receiver points conforms to the survey line design and whether there are any coordinate anomalies or geometric contradictions. The loading profile of the observation system, which has passed quality control, is directly used for subsequent processing, providing an accurate spatial reference frame for steps such as filtering, noise suppression, and direct wave cutoff, ensuring the accuracy and efficiency of the overall imaging process.

[0066] Step 3: Based on the loading of the observation system, frequency-division multichannel statistical anomaly amplitude suppression and high-pass filtering are performed to remove low-frequency strong energy interference, while protecting the effective low-frequency signal and improving the signal-to-noise ratio. After loading the observation system, high-pass filtering is applied to the seismic data. The cutoff frequency of the high-pass filter is set to 5 Hz, and a zero-phase filter design is adopted to minimize the impact of phase distortion on the effective signal. The filtering process suppresses high-frequency environmental noise caused by factors such as ship machinery vibration and cable friction, while ensuring that the effective low-frequency components below 5 Hz are completely preserved. In specific implementation, according to the marine seismic acquisition specifications, the power spectrum of the data is first analyzed to confirm the main energy of the effective signal. The magnitude distribution ranges from 5 to 120 Hz, and a protection frequency band is determined accordingly. During the filtering process, a recursive filtering algorithm is employed to ensure processing efficiency and maintain the stability of the amplitude relationship. This eliminates high-frequency interference and creates conditions for subsequent suppression of low-frequency high-energy noise, avoiding the loss of effective low-frequency information that may occur with conventional broadband filtering. For low-frequency high-energy background noise caused by surges, water flow disturbances, etc., frequency division processing technology is used to decompose the seismic data into three key frequency bands: low-frequency band (5-15 Hz), mid-frequency band (15-45 Hz), and high-frequency band (45-120 Hz). For each frequency band, a dynamic time window length and amplitude threshold are set: the low-frequency band uses... A 200-millisecond time window is used, with the amplitude threshold set to 4.2 times the root-mean-square amplitude within that window. For the mid-frequency band, a 120-millisecond time window is used, with a threshold of 3.8 times the root-mean-square amplitude. For the high-frequency band, an 80-millisecond time window is used, with a threshold of 3.5 times the root-mean-square amplitude. Within each frequency band, using a common receiver gather as the processing unit, multi-channel amplitude statistical characteristics are calculated window-by-window. When the amplitude of a channel exceeds the set threshold and similar anomalies appear in three or more adjacent channels, it is determined to be abnormal amplitude interference. This accurately distinguishes noise characteristics in different frequency ranges and provides accurate location information for subsequent suppression. Based on the identification of abnormal amplitudes, a multi-channel statistical method is used for suppression processing. For the marked abnormal amplitude channels, their amplitude is attenuated to 85% of the threshold. Bidirectional linear interpolation reconstruction is performed using the amplitude values ​​of adjacent normal channels. The interpolation weights are dynamically adjusted based on the channel spacing and similarity coefficient to ensure spatial continuity. For strong energy strip noise with spatiotemporal consistency, an additional tilt scanning technique is used to perform local median filtering along the noise tilt direction within the common offset channel set to further eliminate residual interference. All processing is completed while maintaining phase consistency across frequency bands. The processed data is verified by spectral comparison to ensure that the signal-to-noise ratio improvement in the frequency bands above 5 Hz is no less than 6 dB, and the effective signal amplitude distortion is controlled within 3%. The final output data completely preserves the effective low-frequency reflection information while suppressing strong energy interference.

[0067] Step 4: In response to strong interference from direct waves in shallow water conditions of long cables, accurately cut off the mid-to-long offset direct waves in the filtered data to eliminate interference with ground reflection waves, highlight the effective signal, and ensure that the reflected signal is clear in order to obtain the data after the direct waves are cut off.

[0068] Step 5: Based on the data after direct wave shearing, a precise superimposed velocity field is established using high-order velocity analysis and multiple technical means. Through correction, compatible imaging of reflection interfaces with different tilt angles is achieved, and the superimposed effect is optimized.

[0069] Step 6: Based on the velocity analysis, perform dynamic correction and initial stacking, and implement amplitude energy compensation on the stacked profile to enhance the deep reflection signal. Analyze the amplitude attenuation characteristics, select the optimal parameters to further compensate for the deep reflection energy, enhance the continuity of the same axis, and finally output high-resolution two-dimensional seismic imaging results.

[0070] Example 2, as Figure 1 As shown, based on Embodiment 1, the present invention provides a technical solution: Step 4 specifically includes: In the actual operation of two-dimensional seismic data processing at sea, the identification of direct waves is based on their energy, time-distance curve characteristics, and apparent velocity on the common shot gather. For specific acquisition parameters, the theoretical time-distance relationship of direct waves can be expressed as follows: ,in, This represents the theoretical arrival time of the direct wave on the common shot point gather. This is the offset distance. Because of the water depth, The mean acoustic velocity of the seawater layer is given. In practical processing, the theoretical travel time of the direct wave for each trace is calculated using the precise coordinates of the shot point and receiver. Combined with high-energy, high-continuity in-phase axes from actual seismic records, a semi-automatic tracking method is used on the interactive display profile to confirm the actual distribution range of the direct wave. For long-cable acquisition systems, the time window overlap between the direct wave and shallow reflection waves is significant in the mid-to-long offset region (offset greater than 500 meters), requiring apparent velocity filtering (apparent velocity close to...). Further differentiation is needed; based on the identification results, combined with water depth data, offset distance, and surface velocity model, the direct wave cutoff time window is determined for each channel; for shallow water areas (water depth... (meters), direct wave at offset distance Effective reflection can occur at a distance of 1 meter; in deep water ( Interference mainly occurs at larger offsets (meters). The cutoff window is set to an upper limit of the direct wave arrival time plus a protection window (50 milliseconds), and a lower limit is determined based on the minimum arrival time of the reflected wave field and the seabed reflection time to avoid cutting off effective signals. For rugged seabeds, the cutoff window needs to be smoothly adjusted along the survey line direction to avoid insufficient or excessive cutoff due to seabed undulations. The cutoff parameters are stored in the form of a time window table, including the start and end cutoff times for each shot and each track, to ensure spatial continuity and temporal consistency. The cutoff operation is performed on the pre-processed common shot gather, using a soft cutoff (cosine ramp) method to achieve gradual energy change, with the ramp length set to... 20-40 milliseconds to reduce edge effects, automatically attenuate the amplitude of the direct wave distribution area according to the time window table, and perform energy equalization on the cut gather to avoid abrupt amplitude changes caused by the cut. Quality control is completed in an interactive environment: first, check whether the phase axis of the direct wave on the cut profile is effectively suppressed; second, verify whether the continuity of shallow reflected waves (especially seabed reflections) has been improved; at the same time, ensure that the effective frequency band is not affected by the cut operation through spectrum analysis; the cut data volume is used as the input for velocity analysis and dynamic correction to ensure that the mid-to-long distance offset data is not affected by the residual interference of the direct wave and improve the quality of the stacked imaging.

[0071] Furthermore, step 4 also includes: In actual marine 2D seismic data processing, the determination of the direct wave cutoff range needs to be closely combined with the actual water depth changes and seabed topographic features of the work area. Based on the accurate water depth values ​​of each shot recorded during acquisition (using sonar bathymetry data as input, with an accuracy better than 0.5%), the total cable length, and the real-time offset of each trace, the theoretical arrival time of the direct wave for each seismic trace is dynamically calculated. The core time window calculation formula is as follows: ,in, This is the theoretical arrival time of the direct wave from the seismic source to the detector. Offset distance, i.e., the horizontal distance between the shot point and the detector. The water depth refers to the depth of the seabed at the location of the firing point. The average acoustic velocity in the seawater layer is typically determined after correction for temperature-salinity profiles (typical value is 1500±20 m / s); for shallow water areas (water depth... The minimum interferometric offset threshold is set to 250 meters; when the water depth is between 80 and 150 meters, the minimum interferometric offset threshold is adjusted to 500 meters; when the water depth is greater than 150 meters, the minimum interferometric offset threshold is set to 800 meters. The end time of the time window is increased by a dynamic protection window of 30-80 milliseconds based on the theoretical direct arrival time. The length of this window is adaptively adjusted according to the difference between the two-way travel time and the direct arrival time of the seabed reflected wave. All calculation parameters are stored in the form of a time window table, including the shot number, track number, start cut time (milliseconds), and end cut time (milliseconds). For work areas with severe seabed topographic undulations, spatial adaptive adjustment of the cut parameters is implemented in the common center point (CDP) domain. A seabed undulation surface model is generated based on high-resolution water depth data. Control points are set at intervals of 25 CDPs along the survey line direction. At each control point, the seabed depth change gradient within the range of 50 CDPs before and after it is calculated (the formula is...). , The gradient of seabed depth represents the rate of change of seabed depth per unit distance. To determine the maximum seabed depth within the calculation window, The minimum seabed depth within the calculation window. To calculate the horizontal distance of the window (corresponding to the spatial span of 50 consecutive CDPs), the slope of the cut-off window is dynamically adjusted. When the gradient is greater than 0.15, the surface tracking algorithm is automatically activated to fit the lower boundary of the cut-off window to the time-depth variation of the seabed reflection phase axis. Simultaneously, on the common receiver gather, for sections where the seabed reflection time variation exceeds 40 ms / 100 meters, the width of the cut-off window is narrowed by 15%, and a cosine slope with a length increased to 50 ms is used at the window boundary for transition, ensuring that the effective reflected signal is not excessively cut off in steep seabed areas while maintaining the suppression effect of direct wave interference. All adjustment parameters are recorded in the quality control log for subsequent traceability and verification. Under the premise of meeting geological requirements, an optimized cut-off scheme is provided to improve the efficiency of rapid on-site processing. For seismic traces with an offset greater than 600 meters, the termination time of the standard cut-off window is extended by 15%, i.e., increased by 60- An additional 100 millisecond cutoff is used to ensure that direct wave interference in the mid-to-long offset region is completely suppressed. This is determined by analyzing the interference patterns of direct waves and primary reflected waves in historical data of the work area, and is particularly suitable for work areas with water depths less than 120 meters and relatively flat seabeds. After the cutoff is implemented, three automatic quality checks are immediately performed: First, the energy attenuation of the in-phase axis of the direct wave is verified to be greater than 20 dB on the common shot gather; second, the amplitude continuity loss of seabed reflection is checked to be controlled within 10% on the near offset (<300 meters) gather; and finally, the signal distortion in the 2-150 Hz effective frequency band is confirmed to be less than 3% in the spectral domain. All check results are displayed in real time on the monitoring interface. If the standards are not met, an early warning is triggered and a parameter adjustment plan is recommended. Through this standardized process, the direct wave cutoff processing of a 1000 shot line is completed within 15 minutes, and a high-quality data volume that meets the requirements of subsequent velocity analysis is output.

[0072] Step 5 specifically includes: On the data volume after direct wave shearing, based on seismic geological conditions and imaging requirements, a velocity analysis point is set at intervals of 50 common center points (CDPs) along the survey line direction; in areas with complex geological structures (fault development zones or areas with abrupt changes in strata dip), the analysis point density is increased to one point every 25 CDPs, with each analysis point covering 24 channels before and after it, ensuring sufficient lateral representativeness of the velocity extraction. Higher-order velocity analysis techniques are employed, specifically using the fourth-order dynamic correction formula:

[0073] ;

[0074] in, For seismic waves at a specific offset distance The two-way travel time at a given location represents the total propagation time of the seismic wave from its source, after reflection from the subsurface interface, to the geophone at that offset. This is the observation time that dynamic correction requires. For zero offset time, For the superposition speed, To obtain high-order horizontal velocity parameters, a higher-order correction term is introduced to effectively compensate for time difference distortion caused by large offsets (maximum offset up to 3000 meters), significantly improving the accuracy of deep-sea (round-trip travel time greater than 3 seconds) velocity estimation. During the analysis, velocity spectrum energy clusters at different time windows (intervals of 200 milliseconds) for each analysis point are automatically extracted, and the optimal velocity curve is fitted interactively. After obtaining the velocity functions for each analysis point, velocity spectrum analysis, constant-speed scanning, and variable-speed scanning techniques are used for cross-validation. The velocity spectrum analysis employs 256-channel super-gather calculations and frequency... The velocity range is limited to 5–80 Hz to improve velocity resolution in deep, weak signal areas. Normal-speed scanning generates superimposed profiles at 50 m / s intervals within the range of 1500 m / s to 3500 m / s. Variable-speed scanning, based on a preliminary velocity model, performs local optimization in 5% increments within a ±15% velocity disturbance range. After each scan cycle, the corresponding superimposed profile and isovelocity profile are displayed in real time, and visual comparison is supported in the interactive interface. Operators judge the velocity based on the continuity of in-phase axes in the profile, the degree of fault repositioning, and the convergence of diffracted waves. To verify the model's rationality, inconsistencies in velocity (velocities between adjacent CDPs exceeding 10% or deep velocity reversal regions) were automatically identified, and a preliminary quality control report was generated. Based on the preliminary velocity model and the quality control report, iterative optimization was conducted to establish an accurate superimposed velocity field that adapts to both longitudinal and lateral variations. The optimization process employed a combination of tomographic inversion and inter-layer grid interpolation: the velocity field was divided into multiple velocity layers, with each layer defined as 100 milliseconds; intra-layer lateral smoothing was achieved using the velocity functions of adjacent velocity analysis points through cubic spline interpolation, ensuring that velocity changes were natural and consistent with the terrain. The system automatically initiates local densification analysis for velocity inconsistencies, re-extracts the velocity spectrum at intervals of 10 CDPs, and performs vertical consistency correction in conjunction with upper and lower layer velocity constraints. After each iteration, the superimposed profile and isovelocity profile are regenerated until the overall signal-to-noise ratio of the superimposed profile across the entire survey line is improved by no less than 15%, and the continuity of the structural morphology in the lateral direction meets the requirements for engineering interpretation. The final output velocity field model is stored at a grid density of 25 CDPs × 100 milliseconds, with a confidence index, and is directly used for subsequent dynamic correction and pre-stack time migration processing.

[0075] Step 5 also includes: During velocity spectrum analysis, dense sampling and fine analysis are implemented for areas with weak deep reflection signals and low signal-to-noise ratios. In practice, with a two-way travel time of 3 seconds as the boundary, when the signal-to-noise ratio (SNR) of the deep effective in-phase axis is lower than 1.5, the density of velocity analysis points is increased from one point every 50 CDPs to one point every 25 CDPs, and the coverage of each analysis point is expanded to 36 channels before and after to enhance the lateral representativeness of the data; an automatic energy cluster picking technology based on cross-correlation algorithm is adopted, with a picking time window of 200 milliseconds and a sliding step of 50 milliseconds, to extract the velocity spectrum energy distribution in the 5-80 Hz frequency band; for the blurred areas of deep energy clusters, a Semblan-based method is started. The local optimization algorithm for the CE spectrum performs a second fine scan at 15 meters per second intervals within a velocity search range of 1500 m / s to 3500 m / s to ensure that the velocity picking accuracy meets the requirements of deep imaging. Simultaneously, it automatically records the velocity function confidence index for each analysis point, triggering a manual interactive verification process when the confidence level falls below 0.7. Dynamic correction processing is performed on the common center point (CDP) gather using the established stacked velocity field model. Specifically, a fourth-order dynamic correction algorithm is employed, with input parameters including: zero-offset two-way travel time, offset, stacking velocity, and higher-order horizontal velocity parameters. During processing, for long-array data with a maximum offset of 3000 meters, the dynamic correction amount corresponding to each offset is automatically calculated. Following the industry standard SPS specification requirements; for areas with varying strata dip angles (dip angle > 15°), the dip angle time difference correction module is activated to achieve accurate time difference correction at different dip angle reflection interfaces while maintaining the effective signal amplitude characteristics unchanged. The corrected gathers meet the following quality control standards: the inter-trace time difference residual at near offsets (< 500 meters) is less than 2 milliseconds, and the flatness error of the phase axis of the mid-to-far offset gathers does not exceed 1 / 3 of the sampling interval; after completing the dynamic correction processing, a systematic quality verification is performed, checking the flatness of the phase axis on the common imaging point gathers, and using a multi-channel cross-correlation algorithm to calculate the remaining time difference, requiring the average remaining time difference across the entire survey line to not exceed 4 milliseconds; for complex structural areas (fault development zones), additional [measures are performed]. Local dip scanning verification ensures the consistency of time difference correction for reflected wave groups on the upper and lower walls of the fault. At the same time, spectral analysis verifies the effective frequency band protection, requiring the amplitude spectrum distortion rate to be less than 3% in the 5-120 Hz dominant frequency range. Qualified gathers enter the partial stacking test stage: representative CDP gathers are selected for constant speed scanning tests, with the scanning speed range covering ±10% of the modeling speed and the scanning interval being 20 meters per second. When the continuous length of the main reflecting layer in the test stacking profile increases by more than 15%, and the cross-sectional wave repositioning error is less than 1 / 2 of the CDP spacing, the dynamic correction processing is deemed to have met the standards and is ready to enter the final stacking process. All processing parameters and verification results are recorded in the quality control database, supporting full-process traceability.

[0076] Step 6 specifically includes: After completing the dynamic correction processing, based on the optimized stacking velocity field, initial stacking is performed on the common center point (CDP) gathers of the entire survey line. The stacking process adopts a weighted average algorithm for each trace, with the weights dynamically allocated according to the offset and signal-to-noise ratio of each trace within the CDP gather. The weight of traces with near offsets (less than 500 meters) is set to 1.2, and the weights of traces with medium and far offsets decrease sequentially to 0.8 to balance the contribution of different offsets to the stacking results and suppress residual noise. The stacking time window length is consistent with the dynamic correction time window, using a 200-millisecond sliding time window with a step size of 50 milliseconds to ensure the stability of time-varying stacking. After the initial stacked profile is generated, the overall signal-to-noise ratio of the profile is automatically calculated, requiring an improvement of no less than 15%. Simultaneously, the main... The transverse continuity of the reflecting layer (such as the seabed or strong impedance interfaces) must be maintained, with the proportion of amplitude abrupt changes controlled within 5%. Overlay parameters and intermediate results are recorded in real-time to the quality control log, supporting the tracing and verification of anomalies. A systematic amplitude attenuation characteristic analysis is performed on the initial overlay profile to quantify the energy loss of seismic waves during propagation. The analysis is based on a time-varying amplitude statistical method, using 500 milliseconds as an analysis window for the two-way travel time. Within each window, the root mean square amplitude value is calculated, and an amplitude attenuation curve over time is fitted. In actual analysis, the focus is on identifying amplitude attenuation caused by spherical diffusion (proportional to time), frequency-varying attenuation caused by formation absorption (using a quality factor Q-value model, with the Q-value range set from 80 to 200), and interface attenuation. Transmission loss (calculated based on the acoustic impedance difference model); for deep reflections (round-trip time greater than 3 seconds), the amplitude attenuation rate typically reaches 40% to 60% for every additional second, and the effective frequency band's dominant frequency decreases by approximately 15 Hz. The analysis process automatically marks areas of abnormal amplitude enhancement or attenuation and cross-validates this with velocity fields and geological strata to ensure that the attenuation characteristics conform to actual geological and propagation laws. Based on the amplitude attenuation characteristic analysis results, time-varying and space-varying amplitude energy compensation is implemented using an exponential gain model compensation function, whose gain coefficient is dynamically adjusted with the round-trip time: the compensation coefficient for shallow layers (0-2 seconds) is 1.0 to 1.5, the compensation coefficient for middle layers (2-4 seconds) increases to 2.0 to 3.0, and the compensation coefficient for deep layers (above 4 seconds) can reach 4. The compensation function is shaped in the frequency domain to reduce the amplification of high-frequency noise during the compensation process, limiting the gain amplitude outside the effective frequency band (5-120 Hz) to no more than 20%. After compensation, the energy increase of the deep reflection phase axis needs to reach 200% to 300% of that before compensation, and the amplitude distortion rate of the shallow effective signal is controlled within 3%. The final output is a high-resolution two-dimensional seismic imaging profile, whose full-band signal-to-noise ratio is improved by no less than 25% compared with the initial stacked profile, and the continuous length of the main reflection interface in the lateral direction increases by more than 20%, meeting the accuracy requirements of marine engineering geological interpretation and resource investigation. All compensation parameters and processing results are archived in the project database to ensure the repeatability and verifiability of the processing process.

[0077] Example 3, as Figures 1 to 12 As shown, based on Embodiments 1 and 2, the present invention also provides a technical solution: Step A: raw data merging and decompilation; which involves merging the raw SEGD data of each shot for each survey line into one data, and decompiling it into internal data that can be applied by Echos software, analyzing the track head information of the data, and analyzing whether the recorded acquisition information is correct by comparing it with the acquisition parameters.

[0078] Step B: Load the observation system; assign values ​​to the track head information according to the acquisition parameters. The track head information mainly includes: track number 600, shot distance 37.5m, track spacing 12.5m, sampling rate 2ms, minimum offset distance 170m, and recording length 12030ms. Figure 2 This is the profile after the observation system is loaded.

[0079] Step C: Filtering and noise suppression;

[0080] Abnormal amplitude noise is unavoidable during marine seismic acquisition. Poor sea conditions, strong winds and waves, foreign objects stuck in underwater equipment, and cable leakage are the main causes of this noise. These abnormal amplitude noises are characterized by strong amplitude, low frequency, wide distribution, and longitudinal stripes. The noise on a single shot and a single track profile is similar from beginning to end, without obvious changes, and has consistent characteristics in both time and space.

[0081] The method for suppressing anomalous amplitude noise first analyzes the noise's spectral range, distribution characteristics, and differences in amplitude and energy compared to the effective wave. The processing principle is to maximize the representation of the acquired data while minimizing the loss of low-frequency effective components. Addressing the characteristics of irregular noise such as surge waves—low frequency and high energy—a frequency- and time-division multichannel statistical suppression technique for anomalous amplitude is employed. This technique effectively suppresses anomalous noise while protecting low-frequency components. Specifically, an appropriate time window and threshold are first set. Within the time window of different frequency domains, the amplitude value is compared with the threshold. For channels (traces) with significant differences in amplitude from the threshold, the amplitude value is first attenuated and then interpolated using the normal amplitude value of adjacent channels.

[0082] Figure 3 and Figure 5 These are the noises recorded on a single shot and a single track, respectively. Due to the difference in amplitude energy, there are many strip-shaped abnormal amplitude noises on the profile. Figure 4 and Figure 6 These are the filtered single-shot and single-track profiles, respectively. The comparison between the single-shot and single-track profiles before and after filtering, and the difference profile between the two, are used to analyze the results. Figure 7It can be seen that the selection of filtering methods and parameters is reasonable, abnormal amplitude noise is well suppressed, and effective signals are well protected.

[0083] Step D: Direct wave cutoff; Direct waves are seismic waves that are received directly by the seismic towed cable geophone after passing through the seawater layer from the epicenter. Compared to seismic waves reflected from the seabed interface, direct waves have a shorter propagation path, resulting in lower signal attenuation and higher energy. In cases of long cables and shallow water, the mid-to-long-range paths of the direct waves intersect with seismic waves reflected from below the seabed, causing interference with the effective signal. Therefore, due to the presence of direct waves, a significant portion of the effective signal at long offsets in seismic data acquired by long cables is blocked by the effective signal from the mid-to-shallow layers. Figure 8 Cutting off the direct waves from the mid-to-long-range channels during processing is an effective method to suppress direct wave interference. Figure 9 For direct wave cut-off profiles, marine data processing can increase the cut-off range of seismic traces with longer offsets to achieve higher efficiency and simplicity. It is particularly important to note that the CDP spacing must be appropriate when cutting off rough seabeds with direct waves; otherwise, excessive seabed cut-off may occur.

[0084] Step E: Velocity Analysis; Appropriately select velocity analysis points and parameters. Employ high-order velocity analysis to reduce the impact of velocity variations due to gun-receiver distance on deeper layers. Considering the complex geological structure of this area, the significant variations in velocity both horizontally and vertically, and the weak deep-layer reflection signals and low signal-to-noise ratio, a comprehensive approach is adopted, utilizing velocity spectra, constant velocity scanning, variable velocity scanning, stacked profiles, and isovelocity profiles to establish an accurate stacked velocity field. Velocity analysis ( Figure 10 (a) is the superimposed profile after dynamic correction; (b) is the velocity picking profile. The quality of multichannel seismic profile processing is crucial.

[0085] Step F: Dynamic correction and initial stacking; Dynamic correction is compatible with reflections at different dip angles occurring simultaneously at subsurface interfaces, and is very effective in imaging complex subsurface structures, laying a good foundation for subsequent data stacking. After dynamic correction and velocity analysis, the data can be initially stacked, resulting in a profile after initial stacking ( Figure 11 The data shows that the resolution of the seismic data can be roughly divided into three regions: 0-4s during two-way travel, 4s-6.5s, and below 6.5s. The resolution of the shallow layers is very high, with clear and continuous reflection interfaces and well-defined contact relationships. However, the resolution is relatively lower in the region of approximately 4s-6.5s, which is considered to be related to the shielding effect of strong reflection interfaces on seismic energy in this region. Below 6.5s, the energy is significantly attenuated, the continuity of the phase axis is weak, and the signal-to-noise ratio is the lowest in this data set.

[0086] Step G: Superimposed Profile Amplitude Energy Compensation; During its propagation, the excitation energy from the seismic source is affected by spherical diffusion, formation absorption, interface transmission, and noise systems. This results in losses of both excitation and reflection energy, leading to weak effective reflection energy at deeper layers and poor data quality. To compensate for this loss and obtain and highlight the deep reflection energy, a detailed analysis of the data amplitude energy attenuation is necessary during data processing. Appropriate compensation parameters are selected to compensate for the data amplitude. Amplitude energy compensation is a crucial step in improving the quality of processed data. Figure 12 The superimposed profile after amplitude energy compensation, and Figure 11 The deep reflections in the cross-section are clearer compared to the previous ones.

[0087] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0088] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A rapid imaging method for marine 2D seismic data, characterized in that, Includes the following steps: Step 1: Obtain raw SEGD format seismic data of the target area from the seismic acquisition system, merge the shot data of each survey line and decompile them into internal usable data to obtain the data body to be processed; Step 2: Based on the decompiled data volume to be processed, load the observation system and assign seismic trace parameters according to the actual acquisition parameters to establish an accurate spatial and temporal correspondence. Step 3: Based on the loading of the observation system, perform frequency-division multi-channel statistical anomaly amplitude suppression and high-pass filtering, while protecting the effective low-frequency signal; Step 4: For strong direct wave interference in shallow water conditions of long cables, accurately remove the mid-to-long offset direct waves from the filtered data to obtain the data after direct wave removal, including: Identify and locate the direct wave that propagates directly from the earthquake source through the seawater layer to the seismic tow cable. The direct wave has the characteristics of short propagation path and high energy under shallow water and long cable acquisition conditions. Based on the propagation distance curve of the direct wave, its distribution range in the seismic record is determined, covering its performance in mid- to long-distance gathers; Based on the aforementioned distribution range, the corresponding regions of the filtered data are precisely cut off to eliminate the interference and shielding of direct waves on the ground reflection signals. Step 4 also includes: When determining the direct wave cutoff range, the changes in water depth and the ruggedness of the seabed topography in the survey area are taken into account. Based on the relationship between water depth, cable length and offset, the direct wave cutoff time window is dynamically calculated. For rugged seabed topography, the removal parameters in the common center point gather are adjusted to adjust the removal range, and data after direct wave removal is obtained. The determination of the direct wave cut-off range is based on the actual water depth changes and seabed topographic features of the work area. In the face of work areas with drastic seabed topographic undulations, the cut-off parameters are spatially adaptively adjusted within the common center point domain. A seabed undulation surface model is generated based on high-resolution water depth data. Control points are set at intervals of 25 common center points along the survey line. At each control point, the slope of the cut-off time window is dynamically adjusted according to the seabed depth change gradient within the range of 50 common center points before and after it. When the gradient is greater than a preset threshold, the surface tracking algorithm is automatically activated to fit the lower boundary of the cut-off time window to the time-depth change of the seabed reflection phase axis. At the same time, on the common receiver gather, the width of the cut-off time window will be narrowed for sections where the seabed reflection time change exceeds the limit. Step 5: Based on the data after direct wave shearing, a precise stacking velocity field is established using high-order term velocity analysis and multiple technical means to optimize the stacking effect; Step 6: Based on the velocity analysis, perform dynamic correction and initial stacking, apply amplitude energy compensation to the stacked profile, analyze the amplitude attenuation characteristics, select the optimal parameters to further compensate for deep reflection energy, and finally output high-resolution two-dimensional seismic imaging results.

2. The rapid imaging method for two-dimensional seismic data at sea according to claim 1, characterized in that: Step 1 specifically includes: Raw seismic data of the target area is acquired from the seismic acquisition system. The raw seismic data is in SEGD format and each survey line contains multi-shot data. The SEGD format multi-shot data corresponding to each survey line are merged to generate a continuous data file for a single survey line; The merged continuous data files are decompiled and converted into an internally usable data format. Preliminary verification of the track header information is then performed to form a data body to be processed.

3. The rapid imaging method for two-dimensional seismic data at sea according to claim 2, characterized in that: Step 1 further includes: The merging process is carried out on a survey line basis, reading SEGD data for each shot and splicing them according to the shot sequence to form a single survey line data file. The decompilation process parses the SEGD header and data body of continuous data files and converts them into a format that can be recognized by the internal software. The decompiled data undergoes a path head information integrity check and is compared with the collected parameter records to verify the accuracy of the collected information and confirm that the data is accurate and usable. The collected parameters include shot point coordinates, receiver point coordinates, sampling interval, and recording length.

4. The rapid imaging method for two-dimensional seismic data at sea according to claim 1, characterized in that: Step 2 specifically includes: Based on the data volume to be processed after decompilation, the observation system is loaded according to the actual acquisition scheme. The system assigns values ​​to the trace head information of the seismic data. The key parameters for this assignment include the number of traces, shot distance, trace spacing, sampling rate, minimum offset, and record length. By assigning values, a precise correspondence between the spatial location and recording time of the seismic trace is established, and the loading profile of the observation system is generated.

5. The rapid imaging method for two-dimensional seismic data at sea according to claim 1, characterized in that: Step 3 specifically includes: High-pass filtering is applied to the seismic data after loading the observation system in order to suppress high-frequency noise while protecting the effective low-frequency signal components according to a pre-set cutoff threshold. A frequency division processing strategy is adopted to address the low-frequency, high-energy anomalous amplitude interference caused by the background noise of swell waves, which has spatiotemporal consistency. Within different frequency bands, by setting time windows and amplitude thresholds, multichannel statistical methods are applied to identify and suppress abnormal amplitudes, thereby effectively suppressing regular and irregular strong noise and protecting low-frequency signals.

6. The rapid imaging method for marine two-dimensional seismic data according to claim 1, characterized in that: Step 5 specifically includes: Based on the data after direct wave shearing, velocity analysis points were selected, and a higher-order velocity analysis method was used to perform velocity analysis in order to reduce the impact of large offset velocity changes on deep velocity estimation. The velocity spectrum analysis, constant velocity scanning and variable velocity scanning techniques were comprehensively used, and the corresponding superimposed profiles and isovelocity profiles were used for interactive verification. Through the aforementioned interactive verification and iterative optimization, a superimposed velocity field model adapted to changes in longitudinal and transverse velocities is established.

7. A rapid imaging method for marine two-dimensional seismic data according to claim 6, characterized in that: Step 5 further includes: Key velocity analysis points are selected on the velocity spectrum, and dense sampling and fine analysis are carried out for areas with weak deep reflection signals and low signal-to-noise ratios. The common center point gather data are dynamically corrected using the established superimposed velocity field model, which makes the reflection interfaces with different dip angles in the subsurface compatible when superimposed in the same phase.

8. A rapid imaging method for marine two-dimensional seismic data according to claim 1, characterized in that: Step 6 specifically includes: The dynamically corrected velocity field is applied to the initial superposition of the data to generate an initial superposition profile. The amplitude attenuation characteristics of seismic waves during propagation are analyzed due to spherical diffusion, stratum absorption, interface transmission, and noise, while also focusing on the energy attenuation law of deep reflection. Based on the amplitude attenuation characteristic analysis results, the optimal amplitude compensation parameters are selected to perform amplitude energy compensation on the initial stacked profile in order to enhance the energy of deep reflection signals, and finally output a high-resolution two-dimensional seismic imaging profile.

Images