Method and device for measuring whale form based on unmanned aerial vehicle laser scale

Abstract

The invention provides a whale form measuring method and device based on an unmanned aerial vehicle laser scale, and relates to the technical field of wild animal protection, and the method comprises the steps: fixedly installing a laser scale device on an unmanned aerial vehicle, and starting and correcting the laser scale device; tracking the target whale by using the unmanned aerial vehicle to shoot right above the target whale, and emitting a laser beam by using a laser scale device to obtain a target whale image containing a laser point; inputting the target whale image into a trained whale identification analysis model to obtain initial form data; and performing proportional conversion on the initial form data according to the position information of the laser point to obtain actual form data. According to the method, the form of the cetacean is measured through the unmanned aerial vehicle and the laser scale device, no substantial contact is caused to the wild cetacean, and the method has the advantages of being safe, stable and environmentally friendly.

Description

Technical Field

[0001] The present invention relates to the technical field of wildlife protection, and in particular to a whale morphology measurement method and device based on an unmanned aerial vehicle laser ruler. Background Art

[0002] Nowadays, in the context of ecological protection, cetaceans are indicator species of the quality of the ecological environment. The measurement and acquisition of wild individual body size data plays a vital role in the evaluation of population nutritional health status and species protection. In order to obtain the morphological data of wild cetaceans through video measurement, a reference object of known length must be present in the picture.

[0003] At present, large whales are usually approached by boats due to their large size, long time to stay on the surface to breathe, and relatively slow movements, so that the morphological data of whales can be estimated by taking pictures of the boats as reference. However, small whales, such as the Yangtze River porpoise, have the characteristics of small targets, short breathing time, flexible swimming, and complex living environment. It is extremely challenging to take clear photos of wild individuals, and it is even more difficult to measure the morphology of small wild whales through photos with reference objects. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method and device for measuring the morphology of small whales based on a drone laser ruler. The morphology of whales is measured through drone images without causing any substantial contact with wild whales. It has the advantages of safety, stability and environmental protection.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A cetacean morphological measurement method based on an unmanned aerial vehicle laser ruler comprises:

[0007] A laser ruler device is fixedly installed on the drone, and the laser ruler device is started and calibrated;

[0008] Using the drone to track the target cetacean, to take a photo directly above the target cetacean, and using the laser ruler device to emit a laser beam to obtain an image of the target cetacean including the laser point;

[0009] Inputting the target cetacean image into a trained cetacean recognition analysis model to obtain initial morphological data;

[0010] The initial shape data is proportionally converted according to the position information of the laser point to obtain actual shape data.

[0011] Preferably, after the initial shape data is proportionally converted according to the position information of the laser point to obtain the actual shape data, the method further includes:

[0012] A 3D model of the target whale is constructed according to the target whale image and the corresponding actual morphological data, and the nutritional health status of the target whale is evaluated using the 3D model.

[0013] Preferably, a laser ruler device is fixedly mounted on the back of the drone, comprising:

[0014] The claw-shaped plastic clip structure at the bottom of the laser scale device is clipped onto two sides of the drone to install the laser scale device on the back of the drone.

[0015] Preferably, starting and calibrating the laser ruler device comprises:

[0016] Connect the laser ruler device and the external power port of the drone through a dual type-c port transmission line to ensure the power supply of the laser ruler device;

[0017] Turn on the remote control switch of the laser ruler device, press the start button, and check whether the laser start-up operation of the laser ruler device is normal;

[0018] Use an Allen wrench to loosen and tighten the lever arms of the laser ruler device and adjust the angle so that the beams of the laser points on both sides of the drone are parallel and the laser points on both sides are within the shooting field of view of the drone;

[0019] The laser tube lens of the laser ruler device is twisted to make the spot sizes of the laser points on both sides consistent.

[0020] Preferably, the target cetacean is the Yangtze River porpoise.

[0021] Preferably, the method for constructing the cetacean identification and analysis model comprises:

[0022] Obtain a sample training set and an initial model; the sample training set includes a plurality of historical overhead images of finless porpoises; the framework of the initial model is the yolov8 algorithm framework;

[0023] The initial model is trained using the sample training set to obtain the whale identification and analysis model.

[0024] Preferably, the initial shape data is proportionally converted according to the position information of the laser point to obtain the actual shape data, including:

[0025] Measuring the actual distance between the laser points on both sides of the target cetacean image;

[0026] Determine the image scale based on the conversion principle between the preset pixel length and the actual spacing;

[0027] The initial morphological data is proportionally converted based on the image scale to obtain the actual morphological data.

[0028] A whale morphology measurement device based on an unmanned aerial vehicle laser ruler comprises: an unmanned aerial vehicle and a laser ruler device;

[0029] The unmanned aerial vehicle is used to carry the laser ruler device, and the lens of the unmanned aerial vehicle is directed straight downward and vertically to obtain an image of the target whale containing a laser point; the laser ruler device is used to emit a laser point, and the emitted laser point is located within the shooting field of view of the unmanned aerial vehicle.

[0030] The present invention discloses the following technical effects:

[0031] The present invention provides a method and device for measuring the morphology of whales based on a laser ruler of an unmanned aerial vehicle, including: using a laser ruler device to be fixedly installed on a drone, and starting and calibrating the laser ruler device; using the drone to track a target whale to shoot directly above the target whale, and using the laser ruler device to emit a laser beam to obtain a target whale image containing a laser point; inputting the target whale image into a trained whale recognition analysis model to obtain initial morphological data; and proportionally converting the initial morphological data according to the position information of the laser point to obtain actual morphological data. The present invention uses an unmanned aerial vehicle and a laser ruler device to measure the morphology of whales, does not cause any substantial contact with wild whales, and has the advantages of safety, stability and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments 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 labor.

[0033] Figure 1 A flow chart of a method provided by an embodiment of the present invention;

[0034] Figure 2 A schematic diagram of a technical route provided for an embodiment of the present invention;

[0035] Figure 3 A schematic diagram of an image processing operation method provided by an embodiment of the present invention;

[0036] Figure 4A schematic diagram of the installation effect of the cetacean morphology measurement device provided by an embodiment of the present invention;

[0037] Figure 5 A schematic diagram of the back structure of a cetacean morphology measurement device provided by an embodiment of the present invention;

[0038] Figure 6 A schematic diagram of the side structure of a cetacean morphology measurement device provided by an embodiment of the present invention;

[0039] Figure 7 A schematic structural diagram of a remote controller for a laser ruler device provided in an embodiment of the present invention.

[0040] Explanation of the reference numerals: H1-integral laser device; H2-integral UAV; 1-ABS cover plate; 2-ABS packaging box; 3-wire hole; 4-type-c connection hole; 5-laser line group; 6-line width knob; 7-laser mirror group; 8-ABS base; 9-ABS claw-type clip; 10, packaging box cover M3 hexagon socket screw; 11-height adjustment hole; 12-height adjustment hole M3 hexagon socket screw; 13-M3 hexagon socket hinge; 14-clamp M3 hexagon socket screw; 15-laser tube clamp; 16-remote control antenna; 17-remote control indicator light; 18-remote control power switch slide slot; 19-remote control power switch; 20-remote control laser "on" key; 21-remote control laser "off" key. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The purpose of the present invention is to provide a method and device for measuring the morphology of whales based on a UAV laser ruler. The morphology of whales is measured by an unmanned aerial vehicle and a laser ruler device, which will not cause any substantial contact with wild whales and has the advantages of safety, stability and environmental protection.

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a whale morphology measurement method based on an unmanned aerial vehicle laser ruler, comprising:

[0045] Step 100: using a laser ruler device to be fixedly installed on the drone, and starting and calibrating the laser ruler device;

[0046] Step 200: using the drone to track the target cetacean, to take a photo directly above the target cetacean, and using the laser ruler device to emit a laser beam to obtain an image of the target cetacean including the laser point;

[0047] Step 300: inputting the target cetacean image into a trained cetacean recognition analysis model to obtain initial morphological data;

[0048] Step 400: Scale the initial shape data according to the position information of the laser point to obtain actual shape data.

[0049] like Figure 2 As shown, the specific process of this embodiment is as follows:

[0050] Step 1: Use the ABS claw-shaped clip at the bottom of the laser ruler device to snap the two sides of the clip onto the two sides of the drone and hold it firmly. At this time, the laser ruler device can be firmly installed on the back of the drone.

[0051] Process 2: Use a dual type-c port transmission cable compatible with fast charging protocols such as QC and PD to connect the type-c port of the laser ruler device and the type-c port of the drone's external power supply to ensure that the drone supplies power to the laser ruler device and reduce the weight of the laser ruler device itself;

[0052] Process 3: Install a 9V square battery on the remote controller of the laser ruler device, slide the switch on the side of the remote controller of the laser ruler device upward to "N", then the indicator light of the remote controller lights up, press the laser "on" button of the remote controller, and the laser tubes on both sides emit green circular lasers, and the laser beam ensures that the spot remains unchanged at 10mm within 0-50m;

[0053] Process 4: Twist the laser tube lens to ensure that the laser spot sizes on both sides are consistent. Then use an M3 Allen wrench to loosen and tighten the clamps, hinges, and lever arms of the laser ruler device to adjust the angle to ensure that the laser beams on both sides are parallel. After the drone takes off, when the lens is vertically downward, the laser points on both sides are within the drone's field of view. Then press the laser "off" button on the remote controller to avoid consuming battery power and causing unnecessary laser irradiation when the drone takes off.

[0054] Step 5: Take off the drone, track small cetacean target animals in the wild, and take vertical images directly above the animals at a safe flying distance where the drone has no impact on the animals;

[0055] Process 6: When the animal can be stably photographed, press the laser "on" button on the remote controller of the laser ruler device to turn on the laser beam and take a picture of the animal with two laser dots;

[0056] Process 7: Open the captured image on the computer, capture an image with complete animal morphology and laser dots, open the image using Image J software, select the line connecting the two laser dots on the image, and because the laser beams on both sides are arranged in parallel, the actual length between the two laser dots in the image can be known by measuring the distance between the two light dots at close range. The image scale can be determined based on the conversion principle between pixel length and actual length;

[0057] Specifically, the specific sub-processes in process 7 are as follows Figure 3 As shown, first click "Straight" to delineate the two laser points in the connection diagram; click "Analyze"-"Set Scale", fill in the actual length between the two laser tubes in "Known distance", fill in the corresponding length unit in "Unit of length", select "Global", and click "OK"; click "Analyze"-"Tools"-"Scale Bar", fill in the scale style information such as length, height, font size and color to complete the scale setting; finally, click "Straight" to select the body length, body width or any body position of the animal in the image, and obtain the actual body length, body width and other morphological data of this wild finless porpoise in "length".

[0058] Exemplarily, the conversion principle of this embodiment is: the actual length represented by a unit pixel = the known actual length of the object / the pixel length selected in the figure.

[0059] Process 8: Select the complete body length, body width or any morphological measurement of wild cetaceans in the image, and calculate its actual body length, body width and other morphological data based on the scale.

[0060] Specifically, in process 8 of this embodiment, the complete body length, body width or any morphological measurement of wild cetaceans in the image is selected by using a convolutional neural network (CNN) algorithm to automatically identify and measure the wild Yangtze finless porpoise photos taken by drones. In addition, this embodiment also constructs a 3D model to calculate BMI (weight / body length) 2 ) and other nutritional parameters, and established a nutritional status assessment technology for wild Yangtze finless porpoises, providing reliable data for scientifically evaluating the nutritional health dynamics of natural populations of Yangtze finless porpoises.

[0061] Based on the deep learning model, we developed an intelligent recognition machine and screening program for Yangtze River porpoise images, obtained the porpoise's body parameters through intelligent extraction of porpoise image parameters, simulated and constructed a 3D porpoise model, and finally evaluated the nutritional health of the porpoise by calculating nutritional health indices such as BMI and body fat percentage.

[0062] Specifically, this embodiment uses images of feeding and rescuing finless porpoises to complete the annotation of more than 1,000 finless porpoise images, establish a training set, and establish a finless porpoise image recognition and image screening program through the YOLOv8 algorithm framework. The specific process is as follows:

[0063] Step 1: Currently, Labelme is used to label 1,000 overhead images of finless porpoises taken at the Baiji Museum of the Institute of Hydrobiology, Chinese Academy of Sciences.

[0064] Step 2: Use the yolov8 algorithm framework for training, training epoch = 300, batch size = 32, and select the best weight best.pt.

[0065] Step 3: The yolov8 segment detection model uses the weighted best.pt to detect the body size data of the wild finless porpoise, and stores the results in txt. The txt file records the pixel coordinates of the body parts of the Yangtze River porpoise in the photo.

[0066] Step 4: Based on the above pixel coordinates, divide the pixel unit into a number of small triangles with fixed areas, and calculate the surface area of ​​the finless porpoise by calculating the number of triangles covered (triangle segmentation method).

[0067] Furthermore, this embodiment realizes the extraction of body size data of finless porpoises. Establishing an accurate reference scale is the key issue in extracting accurate body size parameters from images. By using a laser scale, the drone can carry a laser scale, and in the later-shot images, a laser point scale of known length is always included, so that the morphological data of the animal in the image can be measured.

[0068] Furthermore, the image data automatic recognition model built based on the YOLOv8 algorithm framework accurately extracts the body parameters of the Yangtze River porpoise. Currently, the Yangtze River porpoise body contour recognition and segmentation have been achieved through the labeled training set, with an accuracy of 0.92. Subsequent expansion of the labeled training set will be able to improve recognition accuracy and achieve fine segmentation and recognition of the labeled morphology of the Yangtze River porpoise in multiple scenes (such as the wild, rescue scenes, etc.) to obtain important morphological data, evaluate the nutritional health of the animal, and serve as an important image basis for animal 3D modeling.

[0069] Furthermore, this embodiment utilizes the existing stranded cetacean rescue network to collect stranded finless porpoise specimens, and selects fresh and complete specimens for morphological and anatomical measurements. The main data include body length, weight, body circumference, surface area and specific gravity of each tissue and organ. The drainage method is used to obtain the volume parameters of the finless porpoise, and a high-definition camera is used to obtain high-definition images of the finless porpoise in all directions for subsequent 3D model construction. After the data collection is completed, the 3D model of the finless porpoise is constructed using Blender (RRID: SCR_008606, v 2.80) software, and the model accuracy is verified. Based on the above research results, combined with the back photos of the finless porpoise obtained by drone photography and the corresponding morphological data obtained by image measurement, a 3D model of the finless porpoise is simulated and constructed, and then the nutritional health status of the finless porpoise is evaluated through the model and related morphological parameters (BMI index, body fat percentage, etc.).

[0070] This embodiment uses a small aerial photography drone (such as DJI Mavic 3 / Mavic 3pro), installs a laser ruler device, and uses the principle that laser light propagates in a straight line in the air. The lens is pointed straight down and vertically to shoot, and obtains images of wild small cetaceans with two parallel laser beams. Because the two parallel lasers propagate in a straight line in the air, the distance between the two laser points will not change due to the transmission distance. By measuring the distance between the two laser points at a static distance, the actual distance between the two laser points in the captured image can be confirmed, and the scale is determined according to the conversion principle formula of pixel length and actual length, and then the connecting lines of various parts of the animal morphology in the photo are selected to measure the morphological data such as body length and body width.

[0071] A second aspect of this embodiment provides a method and device for measuring the morphology of small cetaceans based on a drone laser ruler, wherein the device comprises:

[0072] An unmanned aerial vehicle is used to carry a laser scale device, with the lens pointing straight down and shooting vertically to obtain images of wild whales including the laser point;

[0073] The laser ruler device, the main material of the device is ABS (acrylonitrile-butadiene-styrene) wire, which is printed by a 3D printer and has good characteristics such as light weight, durability, firmness and temperature resistance. The upper part of the device is a rectangular box body, which is used to encapsulate the laser remote switch control components and the powered circuit board. The flat and long hole on the back of the device is used for the type-c port connection, and the other end of the type-c port is connected to the external power supply end of the drone to ensure the power supply of the laser device. The two circular holes on the back of the device are used for the laser tube routing. The lower part of the device is a rectangular base and a claw-shaped ABS material clip. The claw-shaped clip can stably grasp the two sides of the drone to achieve the overall installation of the laser device. The two sides of the device are L-shaped force arms, and the multiple joints are connected by hexagonal screws to clamp the laser tubes on both sides. At the same time, the angle between the two laser tubes can be adjusted by loosening the joints of the force arm to ensure that the two laser beams are parallel when in use. Rotate the line width knob on the laser tube left and right to adjust the size of the laser spot projected, and the size of the spot on both sides must be consistent. The front end of the laser tube uses a glass lens group to emit a perfect circular light spot with the characteristics of small beam divergence angle, high parallelism, uniform brightness, etc. The light spot can remain unchanged at 20mm at a distance of 0-50 meters; the brand new original imported tube core and Taiwan lens can be continuously turned on for 24 hours, the light spot is clear and bright, and it is soft and not dazzling; the built-in imported IC intelligent driver board can provide a wide range of power supply of 2.8-5.2V; the interior is filled with heat dissipation silicone and AB glue, which can not only fully dissipate heat, but also prevent shock and drop; the outer shell is aviation aluminum plated with insulating paint, which can not only fully dissipate heat, but also prevent shock, drop and static electricity; the wires are resistant to high temperature and corrosion, and are suitable for all-weather and all-environment use.

[0074] As an optional implementation, Figures 4 to 6 As shown, the overall laser device H1 includes a rectangular ABS packaging box 2, an ABS cover plate 1 on the box and M3 hexagon socket screws 10 fixed around the packaging box cover. A wire hole 3 and a type-c connection hole 4 are reserved on the back of the packaging box for routing the laser line group 5 and powering the power interface. The two sides are L-shaped laser tube clamps 15, and the laser tubes on both sides are clamped by M3 hexagon socket screws 14 on the clamps. The height adjustment hole 11 cooperates with the height adjustment hole M3 hexagon socket screw 12 to adjust the height and vertical angle of the clamp and the laser tube. The M3 hexagon socket hinge 13 can adjust the parallel angle of the laser tubes on both sides. The line width knob 6 of the laser tube can be rotated to adjust the size of the laser spot emitted by the laser lens group 7. The bottom of the device is an ABS base 8 and a fixed ABS claw-type clamp 9 for installing the overall laser ruler device on the back of the overall drone H2.

[0075] Furthermore, if Figure 7As shown, the remote controller in this embodiment includes a remote control antenna 16, a remote control indicator light 17, a remote control power switch slide 18, a remote control power switch 19, a remote control laser "on" key 20 and a remote control laser "off" key 21; wherein the remote control power switch slide 18 is provided with a remote control power switch 19, the remote control laser "on" key 20 and the remote control laser "off" key 21 can control the on / off state of the laser device, and the remote control indicator light 17 can display the on / off state of the remote controller.

[0076] The beneficial effects of the present invention are as follows:

[0077] (1) During the measurement process, the method of the present invention will not cause any substantial contact with wild cetaceans, will not change the living habits and activity patterns of wild cetaceans, and will not cause any impact on the natural ecological environment. Compared with the traditional morphological measurement of cetaceans by placing a scale, this method has the advantages of safety, stability and environmental protection.

[0078] (2) The measuring device of the present invention is lightweight, portable and easy to carry. When used for field photography, the laser ruler device can be installed in any weather and water conditions to achieve photogrammetric measurement of the morphological data of small wild cetaceans. Compared with traditional large-scale cetacean measuring devices, the present device has the advantages of convenience, efficiency and speed.

[0079] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0080] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A whale morphology measurement method based on an unmanned aerial vehicle laser ruler, characterized in that: include: A laser ruler device is fixedly installed on the drone, and the laser ruler device is started and calibrated; Using the drone to track the target cetacean, to take a photo directly above the target cetacean, and using the laser ruler device to emit a laser beam to obtain an image of the target cetacean including the laser point; Inputting the target cetacean image into a trained cetacean recognition analysis model to obtain initial morphological data; Proportionally converting the initial shape data according to the position information of the laser point to obtain actual shape data; Starting and calibrating the laser ruler device includes: Connect the laser ruler device and the external power port of the drone through a dual type-c port transmission line to ensure the power supply of the laser ruler device; Turn on the remote control switch of the laser ruler device, press the start button, and check whether the laser start-up operation of the laser ruler device is normal; Use an Allen wrench to loosen and tighten the lever arms of the laser ruler device and adjust the angle so that the beams of the laser points on both sides of the drone are parallel and the laser points on both sides are within the shooting field of view of the drone; Twist the laser tube lens of the laser ruler device to make the spot sizes of the laser points on both sides consistent; The initial shape data is proportionally converted according to the position information of the laser point to obtain actual shape data, including: Measuring the actual distance between the laser points on both sides of the target cetacean image; Determine the image scale based on the conversion principle between the preset pixel length and the actual spacing; The initial morphological data is proportionally converted based on the image scale to obtain the actual morphological data.

2. The whale morphology measurement method based on drone laser ruler according to claim 1 is characterized in that: After the initial shape data is proportionally converted according to the position information of the laser point to obtain the actual shape data, the method further includes: A 3D model of the target whale is constructed according to the target whale image and the corresponding actual morphological data, and the nutritional health status of the target whale is evaluated using the 3D model.

3. The whale morphology measurement method based on drone laser ruler according to claim 1 is characterized in that: The laser ruler device is fixed on the back of the drone, including: The claw-shaped plastic clip structure at the bottom of the laser scale device is clipped onto two sides of the drone to install the laser scale device on the back of the drone.

4. The whale morphology measurement method based on drone laser ruler according to claim 1 is characterized in that: The target cetacean is the Yangtze River porpoise.

5. The whale morphology measurement method based on drone laser ruler according to claim 1 is characterized in that: The method for constructing the cetacean identification and analysis model comprises: Obtain a sample training set and an initial model; the sample training set includes a plurality of historical overhead images of finless porpoises; the framework of the initial model is the yolov8 algorithm framework; The initial model is trained using the sample training set to obtain the whale identification and analysis model.

6. A whale morphology measurement device based on a drone laser ruler, characterized in that: include: Unmanned aerial vehicles and laser ruler devices; The unmanned aerial vehicle is used to carry the laser ruler device, and the lens of the unmanned aerial vehicle is directed straight downward and vertically to obtain an image of the target whale containing a laser point; the laser ruler device is used to emit a laser point, and the emitted laser point is located within the shooting field of view of the unmanned aerial vehicle.

Images