Species detection method, system, electronic device and storage medium

Abstract

The present invention belongs to the field of environmental ecological technology, and in particular, relates to a species detection method, system, electronic device and storage medium. By acquiring the image information of the species to be detected, the image characteristics and length characteristics in the image information of the species to be detected are extracted according to the feature extraction model trained by the neural network, and then the length of the species to be detected is calculated according to the length characteristics, and whether the species to be detected is an invasive species is determined according to the length of the species to be detected, the image characteristics and the preset invasive species database. If it is an invasive species, the number of invasive species can be determined according to the frequency of occurrence of the invasive species, and the invasive species activity area can be determined according to the location where the invasive species is found. The present invention can accurately and quickly identify in real time whether the photographed species is an alien invasive species, which solves the problem of low efficiency and inaccurate identification of invasive species by human eyes in the prior art, thereby protecting the species balance in the monitoring area.

Description

Technical Field

[0001] The present invention belongs to the field of environmental ecological technology, and in particular relates to a species detection method, system, electronic equipment and storage medium. Background Art

[0002] At present, more than 660 alien invasive species have been found in my country, of which the largest proportion comes from America. These invasive species not only disrupt the original ecological balance, but some even replace local species, reduce biodiversity, and pose a serious threat to human health and national food security.

[0003] In order to minimize or avoid the invasion of alien species, various measures have been taken, such as strengthening supervision, risk assessment, public education, physical isolation, chemical control, biological control, legislative measures and scientific research. However, the current detection of invasive species in lakes or reservoirs mainly relies on human eyes and passers-by. In the absence of professional knowledge, this method has low accuracy and efficiency.

[0004] Therefore, how to accurately and quickly identify alien invasive species has become an urgent problem that needs to be solved. Summary of the invention

[0005] The present invention provides a species detection method, system, electronic device and storage medium for accurately and quickly identifying invasive alien species.

[0006] In a first aspect, an embodiment of the present invention provides a species detection method, the method comprising: acquiring image information of a species to be detected; determining the characteristics of the species to be detected based on the image information of the species to be detected and a feature extraction model, the feature extraction model comprising a part feature extraction model and a length feature extraction model, the characteristics of the species to be detected comprising part characteristics of the species to be detected and length characteristics of the species to be detected; determining the species type of the species to be detected based on the part characteristics, length characteristics and a preset database of the species to be detected, the species types comprising invasive species and non-invasive species, the preset database comprising part characteristics of invasive species, length characteristics of invasive species and names of invasive species.

[0007] In a possible implementation, before determining the characteristics of the species to be detected based on the image information of the species to be detected and the feature extraction model, the method also includes: acquiring image information of the invasive species; determining the part feature extraction model and the length feature extraction model based on the image information of the invasive species and a convolutional neural network model, the part feature extraction model being used to extract part features, the part features including fish scale features and fish tail features, and the length feature extraction model being used to extract the head position and the tail position.

[0008] In a possible implementation, the determining the part feature extraction model and the length feature extraction model based on the invasive species image information and the convolutional neural network model includes: determining a training data set, a verification data set, and a test data set based on the invasive species image information and a preset ratio; determining a trained convolutional neural network model based on the training data set and the convolutional neural network model; determining a loss image of the trained convolutional neural network model based on the trained convolutional neural network model and the verification data set; determining the part feature extraction model and the length feature extraction model based on the loss image, the trained convolutional neural network model, and the test data set.

[0009] In a possible implementation, after determining the part feature extraction model and the length feature extraction model based on the invasive species image information and the convolutional neural network model, the method further includes: determining the length feature of the invasive species based on the head position and tail position and a preset formula.

[0010] In a possible implementation, the determining of the species type based on the part characteristics, length characteristics and preset database of the species to be detected includes: determining the species similarity by cosine similarity based on the part characteristics of the species to be detected and the preset database; and determining the species type based on the species similarity, the length characteristics of the species to be detected and the preset database.

[0011] In a possible implementation, the method further includes: if the species to be detected is an invasive species, obtaining location information of the species to be detected; based on the location information of the species to be detected, obtaining the occurrence frequency of the species to be detected; based on the occurrence frequency of the species to be detected, determining a frequency heat map of the species to be detected.

[0012] In a possible implementation, after determining the frequency heat map of the species to be detected based on the frequency of occurrence of the species to be detected, the method further includes: determining the species name of the species to be detected based on the preset database; and determining species removal recommendations based on the species name and the frequency heat map of the species to be detected.

[0013] In a second aspect, an embodiment of the present invention provides a species detection system, the system comprising:

[0014] An acquisition module is used to acquire image information of a species to be detected; a determination module is used to determine the characteristics of the species to be detected based on the image information of the species to be detected and a feature extraction model, wherein the feature extraction model includes a part feature extraction model and a length feature extraction model, and the characteristics of the species to be detected include part characteristics of the species to be detected and length characteristics of the species to be detected; the determination module is also used to determine the species type of the species to be detected based on the part characteristics, length characteristics and a preset database of the species to be detected, wherein the species type includes invasive species and non-invasive species, and the preset database includes part characteristics of invasive species, length characteristics of invasive species and names of invasive species.

[0015] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in the first aspect or any one of the implementation methods thereof is implemented.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect or any one of the implementation methods thereof is implemented.

[0017] In a fifth aspect, an embodiment of the present invention provides a computer program product. When the computer program product runs on an electronic device, the electronic device executes the method described in the first aspect or any one of the implementations thereof.

[0018] The present invention provides a species detection method, the main method of which includes: obtaining image information of species to be detected; determining the characteristics of the species to be detected based on the image information of the species to be detected and a feature extraction model, the feature extraction model includes a part feature extraction model and a length feature extraction model, the characteristics of the species to be detected include the part characteristics of the species to be detected and the length characteristics of the species to be detected; determining the species type of the species to be detected based on the part characteristics, length characteristics and a preset database of the species to be detected, the species types include invasive species and non-invasive species, the preset database includes the part characteristics of invasive species, the length characteristics of invasive species and the name of invasive species. In view of the prior art, it mainly relies on the recognition of human eyes and passers-by to discover invasive species in lakes or reservoirs. This method of identifying alien species has low efficiency and high error rate. The present application compares the characteristics of the species with the preset database data, and can accurately and quickly identify whether the photographed species is an alien species in real time, thereby protecting the species balance in the monitoring area. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of a process for detecting species provided by an embodiment of the present invention;

[0021] Figure 2 A monitoring ship provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of a first-order target detector provided by the present invention;

[0023] Figure 4 A loss graph of the feature extraction model provided by the present invention for the fish tail feature;

[0024] Figure 5 This is a recognition result diagram of the underwater garbage recognition model provided by the present invention;

[0025] Figure 6 A schematic diagram of a spectrometer provided in accordance with an embodiment of the present invention;

[0026] Figure 7 A species detection system provided by an embodiment of the present invention;

[0027] Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0029] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0030] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0031] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]", depending on the context.

[0032] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0033] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0034] Species invasion refers to alien species entering new ecosystems through natural or artificial means, settling and reproducing there. These alien species may cause serious damage to local ecosystems, affect biodiversity, and even have a negative impact on human life.

[0035] Alien species may pose a threat to local biodiversity. When alien species enter a new ecosystem, they may grow rapidly in number due to the lack of natural enemies, occupying a large number of ecological niches and squeezing the living space of local species. For example, the invasion of American leeches seriously threatens the biodiversity of European waters. Alien species may disrupt the local ecological balance. Alien species may compete with local species for food and survival resources, resulting in a decrease in the number of local species or even extinction. For example, after the alien lac insect invaded my country, it multiplied in large numbers, occupied a large number of plant resources, and caused a significant decrease in the number of local insects.

[0036] In addition, the invasion of alien species may also lead to the loss of functions of local ecosystems. For example, alligator gar can destroy biodiversity in aquatic environments.

[0037] Therefore, effective measures need to be taken to identify invasive alien species and control them to maintain the local ecological environment. However, at present, the detection of invasive species in lakes or reservoirs mainly relies on the human eye and passers-by. In the absence of professional knowledge, the accuracy and efficiency of this method are low.

[0038] Based on the above problems, the present application invents a species detection method, which can quickly and accurately identify invasive species in aquatic environments, formulate prevention strategies for the identified invasive species, and thus improve local biodiversity.

[0039] Figure 1 A schematic flow chart of a species detection method provided in one embodiment of the present invention, the method comprising:

[0040] S110, obtaining image information of the species to be detected.

[0041] For example, Figure 2 A monitoring ship is provided for one embodiment of the present invention, and the monitoring ship comprises: a computing unit, a lifting platform, a rotator and a camera. In the middle of the rotator is a mechanical arm, which can pick up samples from the water below the target water area.

[0042] Furthermore, a camera is installed on a lifting platform, and the camera captures image information above and below the water through the lifting platform. The camera takes pictures of underwater fish to obtain image information of the fish, that is, image information of the species to be detected.

[0043] S120, determining the features of the species to be detected based on the image information of the species to be detected and the feature extraction model, the feature extraction model includes a part feature extraction model and a length feature extraction model, and the features of the species to be detected include the part features and the length features of the species to be detected.

[0044] Exemplarily, the fish features in the fish image information are obtained based on the image information of the fish and a part feature extraction model and a length feature extraction model in a computing unit on a monitoring vessel.

[0045] In a possible implementation, before determining the characteristics of the species to be detected based on the image information of the species to be detected and the feature extraction model, it also includes: obtaining the image information of the invasive species; determining a part feature extraction model and a length feature extraction model based on the image information of the invasive species and a convolutional neural network model, the part feature extraction model is used to extract part features, the part features include fish scale features and fish tail features, and the length feature extraction model is used to extract the head position and the tail position.

[0046] For example, before extracting features from fish image information, the part feature extraction model and the length feature extraction model must first be trained to ensure accurate extraction of features from the image information of the fish to be detected.

[0047] Specifically, by inputting the image information of common invasive species in the region, and then training the part feature extraction model and the length feature extraction model, the fish scale features and fish tail features, the head position and the tail position in the invasive species image information are extracted. The invasive species image information can be obtained through online pictures.

[0048] In one possible implementation, a part feature extraction model and a length feature extraction model are determined based on invasive species image information and a convolutional neural network model, including: determining a training data set, a verification data set, and a test data set based on the invasive species image information and a preset ratio; determining a trained convolutional neural network model based on the training data set and the convolutional neural network model; determining a loss value of the trained convolutional neural network model based on the trained convolutional neural network model and the verification data set; and determining a part feature extraction model and a length feature extraction model based on the loss value, the trained convolutional neural network model, and the test data set.

[0049] Exemplarily, image information of invasive species in the region is obtained through network images, and then the image information of invasive species is divided into a training data set, a verification data set, and a test data set.

[0050] Specifically, if 300 images of different types of fish and invasive species of different resolutions are collected, the invasive species image information is divided into three groups, with the training data set accounting for 70%, the verification data set accounting for 20%, and the test data set accounting for 10%.

[0051] As an example, the YOLOv5 architecture can be used to extract the fish scale features, fish tail features, head position, and tail position of the training data set in the invasive species image information. The YOLOv5 architecture is a first-order target detector. Figure 3 The schematic diagram of the first-order target detector provided by the present invention, the first-order target detector consists of three parts: the first-order target detector consists of three parts: the backbone network (Backbone), the network layer (Neck) and the detection head (Head), which is used for prediction. The model backbone network is mainly used to obtain important features from a given image. In YOLOv5, the CSPDarknet-53 pre-trained network is used as the backbone network to extract information features (visual features in the image related to the identification of objects) from the input image. In addition, PANet is used as the network layer to obtain the feature pyramid, and three 1×1 convolutional layers are used as the detection head to predict the bounding box position, score and object category, that is, Figure 4 The YOLOv5 architecture uses a deep neural network to divide the image into a grid and predict bounding boxes and class probabilities simultaneously. YOLOv5 is able to capture fine-grained details and small objects. Its efficient architecture enables it to very accurately detect the various features of fish, such as scale features and tail features, head position, and tail position, even in complex underwater environments or when there are multiple fish species.

[0052] As an example, the part feature extraction model and the length feature extraction model are determined based on the invasive species image information and the convolutional neural network model. The invasive species image information used is 300 invasive species image information of different types of fish and different resolutions collected from the same group.

[0053] Furthermore, the convolutional neural network model trained with the training data set is trained with the validation data set to obtain its loss index and error index, i.e., the loss value. Figure 4 The feature extraction model provided by the present invention is a loss diagram for the fish tail feature, the horizontal axis represents the number of trainings, and the vertical axis represents the loss index or error index. It can be seen that as the number of trainings increases, the loss index or error index gradually decreases. When the loss value reaches the preset threshold, it can be determined that the training is completed, and then the convolutional neural network model after training and the test data set are tested to determine whether the convolutional neural network model after training accurately identifies the fish scale features and fish tail features, head position and tail position in the test data set. If they are all accurately identified, the training is completed, and the part feature extraction model and the length feature extraction model are obtained.

[0054] In a possible implementation, after determining the part feature extraction model and the length feature extraction model based on the invasive species image information and the convolutional neural network model, it also includes: determining the length feature of the invasive species based on the head position and the tail position and a preset formula.

[0055] Exemplarily, a length feature extraction model is obtained through a convolutional neural network model, the head position and tail position in the image information of the invasive species are determined, and the length feature of the invasive species is determined according to the head position and tail position and a preset formula. In order to calculate the length feature of the fish, the provided environment should be equipped with an RGB-D camera, which can not only output the received RGB image, but also output the depth information corresponding to each pixel in the image.

[0056] The specific formula is:

[0057]

[0058] z=d (3)

[0059] Where w' and h' are the width and height of the image, d represents the vertical distance from the camera to the object, and α and β represent the vertical and horizontal viewing angles of the camera, respectively. The z-axis value is obtained by calculating the depth using the depth camera information d.

[0060] Combining formulas (1)-(3), the three-dimensional vector length calculation formula (4) is used to calculate the length of the fish, where l represents the length characteristic of the fish.

[0061]

[0062] S130, determining the species type of the species to be detected based on the part characteristics, length characteristics and a preset database of the species to be detected, wherein the species types include invasive species and non-invasive species, and the preset database includes the part characteristics, length characteristics and names of invasive species of the invasive species.

[0063] In one possible implementation, the species type is determined based on the part characteristics, length characteristics and preset database of the species to be detected, including: determining the species similarity through cosine similarity based on the part characteristics of the species to be detected and the preset database; determining the species type based on the species similarity, the length characteristics of the species to be detected and the preset database.

[0064] Exemplarily, the part feature extraction model and the length feature extraction model trained by the neural network are used to extract the part features of the species to be detected, that is, to extract the fish scale features and fish tail features, the head position and the tail position of the species to be detected.

[0065] Furthermore, after the part features of the species to be detected are extracted, the similarity of the images is found through the cosine similarity method. Cosine similarity is an indicator used to measure the similarity between two vectors. In image processing, it can be used to compare feature vectors extracted from images, such as the part features of fish.

[0066] Specifically, the species type is determined based on the cosine similarity of the scale features of the species to be detected and the scale features of invasive species in a preset database, based on the cosine similarity of the tail features of the species to be detected and the tail features of invasive species in a preset database, and the length features of the species to be detected and the preset database.

[0067] Exemplarily, the cosine similarity between two vectors, i.e., the scale similarity or tail similarity of the species to be detected and the invasive species, is calculated based on the cosine value of the angle between them. It ranges from -1 to 1, where 1 means the vectors are exactly the same, 0 means they are not similar, and -1 means the vectors are completely dissimilar.

[0068] Furthermore, the cosine similarity result is combined with the length characteristics of the species to be detected and the length characteristics of the invasive species to determine whether the species to be detected is an invasive species.

[0069] Specifically, the monitoring ship can combine the fish's characteristics and length characteristics to determine the type of fish, and then compare it with a pre-compiled list of invasive species for each continent or region, that is, a preset database, to see if it belongs to the species on the list. Then it will output whether the detected fish is an invasive species.

[0070] In a possible implementation, if the species to be detected is an invasive species, the location information of the species to be detected is obtained; based on the location information of the species to be detected, the occurrence frequency of the species to be detected is obtained; based on the occurrence frequency of the species to be detected, a frequency heat map of the species to be detected is determined.

[0071] Exemplarily, when the camera captures the image information of the species to be detected, the shooting location and the path of the monitoring ship will be recorded. After detecting the invasive species, the monitoring will mark the location where the invasive species was found. Over time, the monitoring ship will save the average frequency of invasive species appearing in the same location of the water body in its local database. Using the logistic regression method, the number of invasive species can be estimated. By forwarding the name and number of the invasive species, and then determining the frequency heat map of the species to be detected based on the frequency of occurrence of the species to be detected. The computing unit of the monitoring ship can be connected to a web-based artificial intelligence generator to obtain recommendations on how to remove these species.

[0072] In addition, the average frequency of invasive species occurrence can be used to draw species abundance heat maps, showing the density of fish populations in different areas of closed water bodies (lakes, reservoirs, etc.). This can provide a user-friendly interface for staff when dealing with these invasive species.

[0073] As another example, underwater debris can be identified by monitoring vessels as follows:

[0074] In a possible implementation, underwater image information is acquired based on a camera; underwater garbage is determined based on the underwater image information and an underwater garbage recognition model, where the underwater garbage recognition model is obtained through convolutional neural network training.

[0075] In a possible implementation, the method of using a convolutional neural network to train an underwater garbage recognition model is similar to the method of using a convolutional neural network to train a part feature extraction model, which will not be repeated here. Figure 5 This is a diagram of the recognition result of the underwater garbage recognition model provided by the present invention. It can be seen that the underwater garbage recognition model can accurately recognize garbage in the water based on the image information.

[0076] In a possible implementation, after determining the underwater garbage based on the underwater image information and the underwater garbage recognition model, it also includes: determining the location of the underwater garbage based on the positioning system; and determining a garbage distribution heat map of the area to be detected based on the location of the underwater garbage and the area to be detected.

[0077] Specifically, according to the positioning system, after identifying the garbage, the monitoring ship will record its current location and the amount of garbage detected. Then, this information will be sent to the local database, which will use this information to output a real-time heat map.

[0078] As another example, a monitoring ship can be used to test the water quality in the area, as follows:

[0079] Obtain samples to be tested in the target waters.

[0080] The color of the sample to be detected after the reaction is obtained based on the sample to be detected and the detection reagent, and the detection reagent changes color according to the content of nitrate in the sample to be detected.

[0081] Exemplarily, the monitoring vessel detects nitrate concentration using the Lambert-Beer law, under which the absorbed light is equal to the nitrate concentration in the sample to be detected. The nitrate content in the sample to be detected is determined by the change in color of the sample to be detected.

[0082] In one possible implementation, the rotator is used to obtain multiple groups of samples to be tested from different sampling locations in the target area, and place the multiple groups of samples to be tested into different chambers in the rotator respectively, and obtain the colors of the samples to be tested after the reaction based on the samples to be tested and the detection reagents, including: obtaining the colors of the detection samples after the reaction based on the multiple groups of samples to be tested and the detection reagents in different chambers.

[0083] Exemplarily, the robotic arm acquires multiple groups of samples to be tested from different sampling locations in the target area, and distributes the multiple groups of samples to be tested into different chambers in the rotator. The monitoring ship can analyze samples from multiple locations / areas at the same time. During the sample acquisition process, the sampling locations of the test samples are recorded by the positioning system.

[0084] S130, obtaining a spectral photo based on the color of the sample to be detected after the reaction and the camera.

[0085] Specifically, a spectral photo is obtained based on the color of the sample to be tested after the reaction and the camera.

[0086] Exemplarily, the monitoring vessel further comprises a spectrometer, and the spectrometer is manufactured according to the Lambert-Beer law. Figure 6 Schematic diagram of a spectrometer provided in one embodiment of the present invention. First, the sample to be tested absorbs the test reagent, and then the color of the sample to be tested changes according to the nitrate content in the sample to be tested. Then, the light in the spectrometer turns on ( Figure 6 The light source shown in the figure) shines in through the small hole in the sample holder ( Figure 6 Focusing element shown). Depending on the amount of nitrate in the sample water, some of the light will be absorbed by the sample. The light will then pass through another slit ( Figure 6 entrance slit shown), into a mirror ( Figure 6 The optical signal is reflected by a small hole on the spectrometer, and the CCD (Charge-coupled Device) array converts the optical signal into an electrical signal, and then forms a spectral image based on the electrical signal. The camera captures the spectral image to obtain a spectral photo.

[0087] Before the spectrometer analysis, the spectrometer needs to be calibrated to avoid errors during the analysis. Three predetermined spectral samples are used for calibration. After that, the concentration of nitrate is stably calculated using the Lambert-Beer law.

[0088] Furthermore, by distributing multiple groups of samples to be tested into different chambers in the rotator, and then combining the spectrometer and spectral photographs, the monitoring ship can analyze samples from multiple locations at the same time.

[0089] Furthermore, the rotator may be polygonal or circular in shape.

[0090] The water quality of the sample to be tested is determined based on the spectral photograph and the calculation unit.

[0091] Exemplarily, based on the spectral photograph, the photograph is sent to a computing unit in the monitoring ship, and the computing unit determines the concentration of nitrate in the sample to be tested based on the python code, thereby determining the water quality of the sample to be tested.

[0092] Furthermore, through long-term data collection, long short-term memory neural networks are used to predict the long-term health of water bodies and determine when algae will reproduce in large numbers.

[0093] In a possible implementation, the monitoring ship also includes a positioning system. After determining the water quality of the sample to be tested based on the spectral photograph and the calculation unit, it includes: determining the sampling location of the sample to be tested based on the positioning system; determining the water quality distribution of the area to be tested based on the sampling location and the water quality of the sample to be tested.

[0094] For example, based on the sampling locations of the samples to be tested in the area to be tested determined by the positioning system, and then based on the water quality corresponding to the samples to be tested, the water quality distribution of each area in the reservoir or other water environment can be determined, and the water quality distribution heat map can be determined based on the water quality distribution.

[0095] In a possible implementation, the monitoring ship also includes an unmanned driving system, which is used in conjunction with a camera to autonomously plan the monitoring ship to reach a designated waypoint while avoiding obstacles on the driving path.

[0096] Specifically, the monitoring vessel is equipped with a navigation and control system that can autonomously plan and navigate a preset route to a designated location while avoiding obstacles and adapting to changing environmental conditions. Autonomous planning enables the monitoring vessel to navigate to the predetermined location efficiently and safely. Once the monitoring vessel reaches the designated waypoint, it notifies the monitoring system of its arrival. In this way, the system can initiate various operations.

[0097] Furthermore, the monitoring vessel can be equipped with two modes:

[0098] 1. Regular patrol mode: It can monitor at specific points regularly to ensure that water quality remains stable. It provides a systematic way to actively monitor and maintain water quality, and timely intervention measures can be taken to protect a stable and healthy aquatic environment.

[0099] 2. Event-triggered monitoring: It can collect data in a targeted manner when a specific ecological event occurs, and has entered a more severe situation that requires all-weather real-time monitoring. High-density, long-term sampling and analysis are carried out at specific locations to find out the root cause.

[0100] After the samples are collected and analyzed, real-time data transmission is also provided. The monitoring ship can transmit the data back to the monitoring system in real time, so that analysis and decisions can be made quickly based on the collected data.

[0101] Monitoring vessels can provide opportunities and buy time for early prevention by analyzing previously collected data sets and their relationship to overall environmental events. This approach involves collecting and analyzing large amounts of data from different sources, including environmental sensors and historical records.

[0102] Furthermore, advanced analytical techniques such as machine learning and data mining are used to identify patterns, correlations, and anomalies. By mining these insights, we can better understand the relationship between different environmental factors and events.

[0103] The technical solution provided by the present invention obtains the image information of the species to be detected; determines the characteristics of the species to be detected based on the image information of the species to be detected and the feature extraction model, the feature extraction model includes a part feature extraction model and a length feature extraction model, and the characteristics of the species to be detected include the part characteristics of the species to be detected and the length characteristics of the species to be detected; determines the species type based on the cosine similarity of the part characteristics of the species to be detected, the length characteristics of the species to be detected and a preset database, the species type includes invasive species and non-invasive species, and the preset database includes the part characteristics of invasive species, the length characteristics of invasive species and the name of invasive species. It can quickly and accurately determine whether the detected species is an invasive species without the need for manual naked eye detection. The accuracy and efficiency of detection are improved. In addition, the number density and activity area of ​​the species can also be determined for the detected invasive objects, which can effectively solve the problem of alien species invasion.

[0104] Figure 7 The species detection system provided by one embodiment of the present invention only shows the parts related to the embodiment of the present invention for the convenience of description. Figure 7 , the species detection system 700 includes an acquisition module 701 and a determination module 702;

[0105] In one implementation, the apparatus 700 may be used to implement the above Figure 1 For example, the acquisition module 701 is used to implement S110, and the determination module 702 is used to implement S120 and S130.

[0106] The technical solution provided by the present invention obtains the image information of the species to be detected; determines the characteristics of the species to be detected based on the image information of the species to be detected and the feature extraction model, the feature extraction model includes a part feature extraction model and a length feature extraction model, and the characteristics of the species to be detected include the part characteristics of the species to be detected and the length characteristics of the species to be detected; determines the species type based on the cosine similarity of the part characteristics of the species to be detected, the length characteristics of the species to be detected and a preset database, the species type includes invasive species and non-invasive species, and the preset database includes the part characteristics of invasive species, the length characteristics of invasive species and the name of invasive species. It can quickly and accurately determine whether the detected species is an invasive species without the need for manual naked eye detection. The accuracy and efficiency of detection are improved. In addition, the number density and activity area of ​​the species can also be determined for the detected invasive objects, which can effectively solve the problem of alien species invasion.

[0107] Figure 8 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 8 As shown, the electronic device 8 of this embodiment includes: at least one processor 80 ( Figure 8Only one is shown in the figure) a processor, a memory 81, and a computer program 82 stored in the memory 81 and executable on the at least one processor 80, and when the processor 80 executes the computer program 82, the steps in any of the above method embodiments are implemented.

[0108] The electronic device 8 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device may include, but is not limited to, a processor 80 and a memory 81. Those skilled in the art will appreciate that Figure 8 It is only an example of the electronic device 8 and does not constitute a limitation on the electronic device 8. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0109] The processor 80 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0110] In some embodiments, the memory 81 may be an internal storage unit of the electronic device 8, such as a hard disk or memory of the electronic device 8. In other embodiments, the memory 81 may also be an external storage device of the electronic device 8, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 8. Further, the memory 81 may also include both an internal storage unit of the electronic device 8 and an external storage device. The memory 81 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 81 may also be used to temporarily store data that has been output or is to be output.

[0111] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0112] An embodiment of the present invention also provides an electronic device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps of any of the above-mentioned method embodiments when executing the computer program.

[0113] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0114] An embodiment of the present invention provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0115] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / electronic device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

[0116] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0117] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0118] In the embodiments provided by the present invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A species detection method, characterized in that: The method comprises: Obtain image information of species to be detected; Determine the characteristics of the species to be detected based on the image information of the species to be detected and a feature extraction model, wherein the feature extraction model includes a part feature extraction model and a length feature extraction model, and the characteristics of the species to be detected include a part feature of the species to be detected and a length feature of the species to be detected; The species type of the species to be detected is determined based on the part characteristics, length characteristics and a preset database of the species to be detected, wherein the species types include invasive species and non-invasive species, and the preset database includes the part characteristics, length characteristics and names of invasive species.

2. The species detection method according to claim 1, characterized in that: Before determining the features of the species to be detected based on the image information of the species to be detected and the feature extraction model, the method further includes: Obtain image information of invasive species; Based on the invasive species image information and the convolutional neural network model, the part feature extraction model and the length feature extraction model are determined. The part feature extraction model is used to extract part features, including fish scale features and fish tail features. The length feature extraction model is used to extract the head position and the tail position.

3. The species detection method according to claim 2, characterized in that: The determining of the part feature extraction model and the length feature extraction model based on the invasive species image information and the convolutional neural network model includes: Based on the invasive species image information and a preset ratio, determining a training data set, a verification data set, and a test data set; Determine a trained convolutional neural network model based on the training data set and the convolutional neural network model; Determining a loss image of the trained convolutional neural network model based on the trained convolutional neural network model and the validation data set; Based on the loss image, the trained convolutional neural network model and the test data set, the part feature extraction model and the length feature extraction model are determined.

4. The species detection method according to claim 3, characterized in that: After determining the part feature extraction model and the length feature extraction model based on the invasive species image information and the convolutional neural network model, the method further includes: The length characteristics of the invasive species are determined based on the head position and the tail position and a preset formula.

5. The species detection method according to claim 1, characterized in that: The method of determining the species type based on the part characteristics, length characteristics and a preset database of the species to be detected includes: Based on the part characteristics of the species to be detected and the preset database, determining the species similarity by cosine similarity; The species type is determined based on the species similarity, the length characteristics of the species to be detected and a preset database.

6. The species detection method according to claim 1, characterized in that: The method further comprises: If the species to be detected is an invasive species, obtaining location information of the species to be detected; Based on the location information of the species to be detected, obtaining the occurrence frequency of the species to be detected; Based on the occurrence frequency of the species to be detected, a frequency heat map of the species to be detected is determined.

7. The species detection method according to claim 6, characterized in that: After determining the frequency heat map of the species to be detected based on the occurrence frequency of the species to be detected, the method further includes: Based on the preset database, determining the species name of the species to be detected; A species removal recommendation is determined based on a frequency heat map of the species name and the species to be detected.

8. A species detection system, characterized in that: The system comprises: An acquisition module is used to acquire image information of species to be detected; A determination module, used to determine the characteristics of the species to be detected based on the image information of the species to be detected and a feature extraction model, wherein the feature extraction model includes a part feature extraction model and a length feature extraction model, and the characteristics of the species to be detected include a part feature of the species to be detected and a length feature of the species to be detected; The determination module is also used to determine the species type of the species to be detected based on the part characteristics, length characteristics of the species to be detected and a preset database, wherein the species types include invasive species and non-invasive species, and the preset database includes the part characteristics of invasive species, the length characteristics of invasive species and the names of invasive species.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images