A monomer target identification method and system for marine organisms and fishery resources
Patent Information
- Application Number
- CN202311538412.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-17
- Publication Date
- 2026-10-27
- Estimated Expiration
- 2043-11-17
AI Technical Summary
但窄带系统在功率一定情况下，存在距离分辨率与作用距离不可兼得的缺陷，并存在回波信息模糊、识别准确率低等问题，对于相距较近的目标产生的重叠回波可能误识别为单体目标，造成相关研究结果出现较大偏差
[0021]本发明公开一种面向海洋生物及渔业资源的单体目标识别方法及系统，对初始水声回波信号进行带通滤波，以得到待处理水声回波信号，然后确定各个回波信号的信号峰值、相邻回波信号峰值间距，进而基于各个回波信号的信号峰值、相邻回波信号峰值间距，确定各个回波信号的类型，包括重叠回波及疑似单体目标回波；基于Hilbert变换，确定各个回波信号的瞬时频率方差，进而对疑似单体目标回波进行类型更新；最后，基于预设脉宽阈值范围，对更新后的疑似单体目标回波进行脉宽判断，并当脉宽处于预设脉宽阈值范围时，将更新后的疑似单体目标回波标记为最终单体目标回波。本发明通过多次识别，能够得到更为准确的单体目标回波。
Smart Images
Figure CN117665827B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of information processing technology, and in particular to a method and system for identifying individual targets in marine organisms and fishery resources. Background Technology
[0002] In marine biological and fisheries resource exploration, underwater acoustic methods and underwater acoustic signal processing technology are among the most important technical means. Underwater acoustics, as the only underwater medium capable of transmitting information over long distances, allows for marine biological exploration and fisheries research based on the acoustic echoes of targets. The detection and identification of echoes from biological targets such as fish is a fundamental technology for conducting biological spectrum analysis, target intensity measurement, tracking, and subsequently, species identification, fisheries resource assessment, and farmed fish monitoring. This is of great significance for reducing by-catch, lowering the rate of accidental catch of juvenile fish, and monitoring and assessing the growth status of farmed fish.
[0003] In current practice, underwater acoustic detection systems based on narrowband signals are widely used, leading to the development of single-target identification techniques based on narrowband signal characteristics such as pulse width and amplitude. These techniques filter overlapping echo signals by setting normalized pulse width and amplitude thresholds to detect single targets. However, narrowband systems, under certain power conditions, suffer from a trade-off between range resolution and effective range, and also exhibit problems such as fuzzy echo information and low recognition accuracy. Overlapping echoes from closely spaced targets may be misidentified as single targets, causing significant deviations in related research results. Summary of the Invention
[0004] The purpose of this invention is to provide a method and system for identifying individual targets in marine organisms and fishery resources, thereby improving the accuracy of identifying individual target echoes.
[0005] To achieve the above objectives, the present invention provides the following solution:
[0006] In a first aspect, the present invention provides a method for identifying individual targets in marine organisms and fishery resources, comprising:
[0007] The initial underwater acoustic echo signal is bandpass filtered to obtain the underwater acoustic echo signal to be processed; the initial underwater acoustic echo signal is obtained by detecting marine organisms and fishery resources using an underwater acoustic detection system;
[0008] Determine the peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed;
[0009] The type of each echo signal is determined based on the signal peak value of each echo signal and the peak value spacing between adjacent echo signals; the types include overlapping echoes and suspected single target echoes.
[0010] Based on the Hilbert transform, the instantaneous frequency variance of each echo signal is determined according to the underwater acoustic echo signal to be processed.
[0011] Based on the instantaneous frequency variance of each echo signal and the preset frequency variance threshold range, the type of the suspected single target echo is updated.
[0012] Based on a preset pulse width threshold range, the pulse width of the updated suspected single target echo is determined, and when the pulse width is within the preset pulse width threshold range, the updated suspected single target echo is marked as the final single target echo.
[0013] Secondly, the present invention provides a single target identification system for marine organisms and fishery resources, comprising:
[0014] The echo signal acquisition module is used to perform bandpass filtering on the initial underwater acoustic echo signal to obtain the underwater acoustic echo signal to be processed; the initial underwater acoustic echo signal is obtained by detecting marine organisms and fishery resources using an underwater acoustic detection system;
[0015] The peak data calculation module is used to determine the signal peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed.
[0016] The initial identification module is used to determine the type of each echo signal based on the signal peak value of each echo signal and the peak value spacing between adjacent echo signals; the types include overlapping echoes and suspected single target echoes;
[0017] The variance data calculation module is used to determine the instantaneous frequency variance of each echo signal based on the underwater acoustic echo signal to be processed, according to the Hilbert transform.
[0018] The secondary identification module is used to update the type of the suspected single target echo based on the instantaneous frequency variance of each echo signal and a preset frequency variance threshold range.
[0019] The final identification module is used to determine the pulse width of the updated suspected single target echo based on a preset pulse width threshold range, and when the pulse width is within the preset pulse width threshold range, the updated suspected single target echo is marked as the final single target echo.
[0020] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0021] This invention discloses a method and system for identifying single targets in marine organisms and fishery resources. The method involves bandpass filtering an initial underwater acoustic echo signal to obtain the signal to be processed. Then, the peak value of each echo signal and the distance between peak values of adjacent echo signals are determined. Based on these factors, the type of each echo signal is determined, including overlapping echoes and suspected single target echoes. The instantaneous frequency variance of each echo signal is determined using Hilbert transform, and the type of suspected single target echoes is updated. Finally, based on a preset pulse width threshold range, the pulse width of the updated suspected single target echoes is judged. When the pulse width falls within the preset threshold range, the updated suspected single target echo is marked as the final single target echo. This invention, through multiple identifications, can obtain more accurate single target echoes. Attached Figure Description
[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 This is a flowchart illustrating the single target identification method for marine organisms and fishery resources according to the present invention.
[0024] Figure 2 This is a schematic diagram of the bandpass filtering of the echo signal according to the present invention;
[0025] Figure 3 This is a schematic diagram illustrating the types of echo signal identification according to the present invention;
[0026] Figure 4 This is a schematic diagram illustrating the present invention's method for identifying single fish echoes based on instantaneous frequency variance and pulse width.
[0027] Figure 5 This is a schematic diagram of the single target identification system for marine organisms and fishery resources of the present invention. Detailed Implementation
[0028] 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.
[0029] The application of broadband underwater acoustic signal detection technology and instruments provides a foundation for improving the detection performance of individual fish. Broadband signal systems possess a large time-bandwidth product, and based on matched filtering techniques, high distance or frequency resolution can be achieved, which helps distinguish nearby targets and detect small targets, making them suitable for single fish echo identification in high-density fish school scenarios. Furthermore, the frequency of broadband signals is time-varying, a time-frequency characteristic not found in narrowband signals, providing a new approach for single fish identification methods based on broadband signal systems. However, it is necessary to identify the characteristics of broadband echo signals from individual fish targets and develop single fish target identification methods suitable for broadband systems.
[0030] Based on the above, this invention provides a method and system for identifying individual targets in marine organisms and fishery resources. It adopts a joint identification method to compensate for echo misjudgment caused by peak detection failure in single detection, while also taking into account applications in low signal-to-noise ratio scenarios.
[0031] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0032] Example 1
[0033] like Figure 1 As shown, this invention provides a method for identifying individual targets in marine organisms and fishery resources, comprising:
[0034] Step 100: Bandpass filtering is applied to the initial underwater acoustic echo signal to obtain the underwater acoustic echo signal to be processed, thereby reducing the interference of noise and reverberation on identification; the initial underwater acoustic echo signal is obtained by detecting marine life and fishery resources using an underwater acoustic detection system. Specifically, a fourth-order Butterworth filter is applied to the initial underwater acoustic echo signal, from which... Figure 2 It can be seen that the signal-to-noise ratio of the signal is improved after filtering.
[0035] Step 200: Determine the peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed; specifically, perform matched filtering on the underwater acoustic echo signal to be processed to obtain the output envelope of the matched filtered signal; based on the output envelope of the matched filtered signal, use a peak detection algorithm to determine the peak value of each echo signal and the peak value spacing between adjacent echo signals.
[0036] In a specific example, based on the output envelope of the matched filter signal, a peak detection algorithm is used to obtain the time corresponding to each peak signal, and the time delay difference between peak times is calculated, which is the peak spacing between adjacent echo signals.
[0037] Step 300: Based on the signal peak value of each echo signal and the peak spacing between adjacent echo signals, determine the type of each echo signal; the types include overlapping echoes and suspected single-target echoes. Specifically, use the peak value threshold obtained from the target intensity of the single-target fish echo and the spacing threshold between peak points obtained from the transmitted pulse width to perform the first identification of the echo signals. Those meeting the threshold requirements are classified as suspected single-target echoes, and those not meeting the threshold conditions are classified as overlapping echoes.
[0038] Step 300 specifically includes:
[0039] (1) Determine whether the peak value of the echo signal is within the preset peak value threshold range, and determine whether the peak value spacing between adjacent echo signals corresponding to the echo signal is within the preset spacing threshold range.
[0040] (2) When the peak value of the echo signal is within a preset peak value threshold range, and the peak value spacing between adjacent echo signals corresponding to the echo signal is within a preset spacing threshold range, the type of the echo signal is a suspected single target echo.
[0041] (3) When the peak value of the echo signal is not within the preset peak value threshold range, or when the peak value spacing between adjacent echo signals corresponding to the echo signal is not within the preset spacing threshold range, the type of the echo signal is an overlapping echo.
[0042] In a specific instance, such as Figure 3 As shown, based on the time delay between the peak points of the signal envelope after matched filtering (i.e., the distance between the peak points of adjacent echo signals), echoes that are close to each other (such as...) are removed. Figure 3 As shown in box A), these are identified as overlapping echoes and discarded. However, when the echoes are far apart, they are sometimes misidentified as single fish echoes (e.g., Figure 3 (As shown in box B). At this time, the signals within boxes B and C are suspected single-target echo signals. The time of the suspected signal is determined by its peak point.
[0043] Step 400: Based on the Hilbert transform, determine the instantaneous frequency variance of each echo signal according to the underwater acoustic echo signal to be processed; specifically, perform the Hilbert transform on the underwater acoustic echo signal to be processed to obtain the instantaneous frequency of each echo signal; and determine the instantaneous frequency variance of each echo signal based on the instantaneous frequency of each echo signal.
[0044] Step 500: Based on the instantaneous frequency variance of each echo signal and a preset frequency variance threshold range, the type of the suspected single target echo is updated. This step is a secondary discrimination based on frequency; those that meet the threshold proceed to the next discrimination stage, while those that do not meet the threshold are excluded as overlapping echoes.
[0045] Step 500 specifically includes:
[0046] (1) For the echo of a suspected single target, the instantaneous frequency variance is obtained based on the instantaneous frequency variance of each echo signal.
[0047] (2) When the instantaneous frequency variance corresponding to the suspected single target echo is within the preset frequency variance threshold range, the type of the echo signal remains unchanged and is still a suspected single target echo.
[0048] (3) When the instantaneous frequency variance corresponding to the suspected single target echo is not within the preset frequency variance threshold range, the type of the echo signal is updated to overlapping echo, that is, from suspected single target echo to overlapping echo.
[0049] The formula for calculating the preset frequency variance threshold range is as follows:
[0050]
[0051] PeakVar represents the limit of the preset frequency variance threshold range, used to distinguish target echoes from noise signals; PeakVAR is only related to the signal-to-noise ratio; α SNR This represents the preset empirical coefficients, specifically the empirical coefficients obtained from tests with different signal-to-noise ratios; T represents the pulse width of the underwater acoustic echo signal to be processed, and β... var (t) represents the instantaneous frequency variance at time t. When the instantaneous frequency variance β var When the frequency variance exceeds a preset threshold, overlapping echoes in the suspected target signal segment are excluded, such as... Figure 4 As shown in box A, this overlapping echo can be effectively eliminated through secondary filtering.
[0052] Step 600: Based on a preset pulse width threshold range, determine the pulse width of the updated suspected single target echo, and when the pulse width is within the preset pulse width threshold range, mark the updated suspected single target echo as the final single target echo.
[0053] Specifically, a preset pulse width threshold range, PeakT, is set to determine the single target. PeakT is only related to the transmitted signal pulse width T. Although the non-overlapping portion is also a diagonal line in the time-frequency domain, its length on the time axis should be less than the single target echo. The single target can be determined using the transmitted pulse time. Figure 4 Box C can be ruled out as an overlapping echo.
[0054] In a specific example, MATLAB 2021b can be used to simulate the single target identification method for marine organisms and fishery resources of this invention.
[0055] In summary, this invention provides a method for identifying single targets based on the combined pulse delay and instantaneous frequency characteristics of broadband underwater acoustic echo matched filtering. When a sonar emits a linear frequency modulated (LFM) signal, overlapping echoes are generated due to the randomness of the spatial distribution of individual fish. The duration of the overlapping portion is expressed as the emitted pulse width minus the time delay difference between the individual echo signals. After matched filtering, the spacing between peak points indirectly reflects the duration of the overlapping portion. Therefore, the peak point delay output by the echo matched filter can be used as one of the conditions for determining whether the echoes overlap. Furthermore, for LFM signals, the echo of a single target maintains time-frequency characteristics similar to the emitted pulse. Theoretically, in the time-frequency domain, it should be a straight line with a slope of the frequency modulation slope k, and its instantaneous frequency variance should theoretically tend to 0. For overlapping echoes, their phase and amplitude exhibit randomness, and their instantaneous frequency variance differs significantly from that of the echo of a single target. Therefore, the time-frequency characteristics of the echo signal can be used as another basis for distinguishing between overlapping echoes and single echoes. Therefore, the present invention uses a joint identification method to compensate for echo misjudgment caused by peak detection failure in single detection, while also taking into account applications in low signal-to-noise ratio scenarios.
[0056] Example 2
[0057] like Figure 5 As shown, in order to implement the technical solution in Embodiment 1 and achieve the corresponding functions and technical effects, this embodiment also provides a single target identification system for marine organisms and fishery resources, including:
[0058] The echo signal acquisition module is used to perform bandpass filtering on the initial underwater acoustic echo signal to obtain the underwater acoustic echo signal to be processed; the initial underwater acoustic echo signal is obtained by using an underwater acoustic detection system to detect marine organisms and fishery resources.
[0059] The peak data calculation module is used to determine the signal peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed.
[0060] The initial identification module is used to determine the type of each echo signal based on the signal peak value of each echo signal and the peak value spacing between adjacent echo signals; the types include overlapping echoes and suspected single target echoes.
[0061] The variance data calculation module is used to determine the instantaneous frequency variance of each echo signal based on the underwater acoustic echo signal to be processed, according to the Hilbert transform.
[0062] The secondary identification module is used to update the type of the suspected single target echo based on the instantaneous frequency variance of each echo signal and a preset frequency variance threshold range.
[0063] The final identification module is used to determine the pulse width of the updated suspected single target echo based on a preset pulse width threshold range, and when the pulse width is within the preset pulse width threshold range, the updated suspected single target echo is marked as the final single target echo.
[0064] Example 3
[0065] This embodiment provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor runs the computer program to enable the electronic device to execute the single target identification method for marine organisms and fishery resources described in Embodiment 1. Optionally, the electronic device may be a server.
[0066] In addition, embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the single target identification method for marine organisms and fishery resources of Embodiment 1.
[0067] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.
[0068] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A method for identifying individual targets in marine organisms and fishery resources, characterized in that the method... include: The initial underwater acoustic echo signal is bandpass filtered to obtain the underwater acoustic echo signal to be processed; The initial underwater acoustic echo signal was obtained by using an underwater acoustic detection system to detect marine organisms and fishery resources; Determine the peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed; The type of each echo signal is determined based on the signal peak value of each echo signal and the peak value spacing between adjacent echo signals; the types include overlapping echoes and suspected single target echoes. Based on the Hilbert transform, the instantaneous frequency variance of each echo signal is determined according to the underwater acoustic echo signal to be processed. Based on the instantaneous frequency variance of each echo signal and the preset frequency variance threshold range, the type of the suspected single target echo is updated. Based on a preset pulse width threshold range, the pulse width of the updated suspected single target echo is determined, and when the pulse width is within the preset pulse width threshold range, the updated suspected single target echo is marked as the final single target echo.
2. The method for identifying individual targets of marine organisms and fishery resources according to claim 1, characterized in that, Bandpass filtering is performed on the initial underwater acoustic echo signal, specifically including: The initial underwater acoustic echo signal was subjected to a fourth-order Butterworth filter.
3. The method for identifying individual targets of marine organisms and fishery resources according to claim 1, characterized in that, Determining the peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed specifically includes: The underwater acoustic echo signal to be processed is subjected to matched filtering to obtain the output envelope of the matched filtered signal; Based on the output envelope of the matched filter signal, a peak detection algorithm is used to determine the signal peak value of each echo signal and the peak value spacing between adjacent echo signals.
4. The method for identifying individual targets of marine organisms and fishery resources according to claim 1, characterized in that, Based on the peak value of each echo signal and the peak value spacing between adjacent echo signals, the type of each echo signal is determined, specifically including: Determine whether the peak value of the echo signal is within a preset peak value threshold range, and determine whether the peak value spacing between adjacent echo signals corresponding to the echo signal is within a preset spacing threshold range; When the peak value of the echo signal is within a preset peak value threshold range, and the peak value spacing between adjacent echo signals corresponding to the echo signal is within a preset spacing threshold range, the type of the echo signal is a suspected single target echo. When the peak value of the echo signal is not within the preset peak value threshold range, or when the peak value spacing between adjacent echo signals corresponding to the echo signal is not within the preset spacing threshold range, the type of the echo signal is an overlapping echo.
5. The method for identifying individual targets of marine organisms and fishery resources according to claim 1, characterized in that, Based on the Hilbert transform, the instantaneous frequency variance of each echo signal is determined according to the underwater acoustic echo signal to be processed, specifically including: The underwater acoustic echo signal to be processed is subjected to Hilbert transform to obtain the instantaneous frequency of each echo signal; The instantaneous frequency variance of each echo signal is determined based on the instantaneous frequency of each echo signal.
6. The method for identifying individual targets of marine organisms and fishery resources according to claim 1, characterized in that, Based on the instantaneous frequency variance of each echo signal and a preset frequency variance threshold range, the type of the suspected single target echo is updated, specifically including: For echoes of suspected single targets, the corresponding instantaneous frequency variance is obtained based on the instantaneous frequency variance of each echo signal. When the instantaneous frequency variance corresponding to the suspected single target echo is within the preset frequency variance threshold range, the type of the echo signal remains unchanged; When the instantaneous frequency variance corresponding to the suspected single target echo is not within the preset frequency variance threshold range, the type of the echo signal is updated to overlapping echo.
7. The method for identifying individual targets of marine organisms and fishery resources according to claim 1, characterized in that, The formula for calculating the preset frequency variance threshold range is as follows: Where PeakVar represents the limit of the preset frequency variance threshold range, α SNR β represents the preset empirical coefficient, T represents the pulse width of the underwater acoustic echo signal to be processed, and β represents the pulse width of the underwater acoustic echo signal to be processed. var (t) represents the instantaneous frequency variance at time t.
8. A single target identification system for marine organisms and fishery resources, characterized in that the system... include: The echo signal acquisition module is used to perform bandpass filtering on the initial underwater acoustic echo signal to obtain the underwater acoustic echo signal to be processed. The initial underwater acoustic echo signal was obtained by using an underwater acoustic detection system to detect marine organisms and fishery resources; The peak data calculation module is used to determine the signal peak value of each echo signal and the peak value spacing between adjacent echo signals in the underwater acoustic echo signal to be processed. The initial identification module is used to determine the type of each echo signal based on the signal peak value of each echo signal and the peak value spacing between adjacent echo signals; the types include overlapping echoes and suspected single target echoes; The variance data calculation module is used to determine the instantaneous frequency variance of each echo signal based on the underwater acoustic echo signal to be processed, according to the Hilbert transform. The secondary identification module is used to update the type of the suspected single target echo based on the instantaneous frequency variance of each echo signal and a preset frequency variance threshold range. The final identification module is used to determine the pulse width of the updated suspected single target echo based on a preset pulse width threshold range, and when the pulse width is within the preset pulse width threshold range, the updated suspected single target echo is marked as the final single target echo.