Forest logging detection methods, devices, electronic equipment and storage media
By installing detection equipment on an outdoor tower and using deep learning algorithms to detect logging features in forest area images, the real-time performance and cost issues of existing forest logging detection methods have been solved, achieving efficient and low-cost forest logging monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA TOWER CO LTD
- Filing Date
- 2022-05-16
- Publication Date
- 2026-05-26
AI Technical Summary
Existing methods for detecting logging in forest areas suffer from poor real-time performance and high detection costs.
By installing detection equipment on an outdoor tower to collect images of forest areas, and using deep learning algorithms to detect the characteristics of logging equipment and tree felling in the images, timely identification and large-scale detection of logging activities in forest areas can be achieved.
It improves the real-time performance of logging detection and reduces detection costs, enabling timely detection of logging activities in forest areas and covering a large detection area.
Smart Images
Figure CN115147716B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, device, electronic device and storage medium for detecting logging in forest areas. Background Technology
[0002] Currently, forest logging detection methods mainly fall into two categories: First, obtaining information on changes in forest vegetation cover through remote sensing satellite imagery. This method can only detect large-scale logging and has poor real-time performance. Second, detecting tree conditions by setting up sensors throughout the forest area. This method requires deploying sensors in various locations within the forest, resulting in high detection costs. It is evident that existing forest logging detection methods suffer from poor real-time performance and high detection costs. Summary of the Invention
[0003] This application provides a method, apparatus, electronic device, and storage medium for detecting logging in forest areas, in order to solve the problems of poor real-time performance and high detection costs in existing logging detection methods.
[0004] The embodiments of this application are implemented as follows:
[0005] In a first aspect, embodiments of this application provide a method for detecting deforestation in forest areas, including:
[0006] If a first feature is detected in a first forest area image, the presence of a second feature in a second forest area image is detected. The first forest area image is acquired by a detection device installed on an outdoor tower. The second forest area image includes all or part of the features of the first forest area image. The second forest area image includes the first feature. The first feature is a feature that matches the features of logging equipment. The second feature is a feature that matches the features of tree felling.
[0007] If the second feature exists in the second forest area image, it is determined that there is logging activity in the forest area corresponding to the second forest area image.
[0008] Secondly, embodiments of this application provide a forest logging detection device, comprising:
[0009] The first detection module is used to detect whether a second feature exists in a second forest area image when a first feature is detected in a first forest area image. The first forest area image is acquired by a detection device installed on an outdoor tower. The second forest area image includes all or part of the features of the first forest area image. The second forest area image includes the first feature, which is a feature that matches the features of logging equipment. The second feature is a feature that matches the features of tree felling.
[0010] The determination module is used to determine that there is logging activity in the forest area corresponding to the second forest area image when the second feature exists in the second forest area image.
[0011] Thirdly, embodiments of this application also provide an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the forest logging detection method described in the first aspect.
[0012] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the forest logging detection method described in the first aspect.
[0013] In this embodiment, a detection device installed on an outdoor tower collects images of forest areas. If features matching logging equipment are detected in the forest area image, further detection is performed to determine if features matching tree felling characteristics exist. This allows for the determination of whether logging activity has occurred in the corresponding forest area. In this embodiment, since outdoor towers are typically located in high-altitude forestry areas, the detection device installed on the tower can detect the area from top to bottom, covering a large detection area and enabling wide-area forest area detection at a low cost. Furthermore, feature detection of the forest area images allows for timely detection of logging activities, improving the real-time performance of logging detection. Attached Figure Description
[0014] Figure 1 This is a flowchart of the forest logging detection method provided in the embodiments of this application;
[0015] Figure 2 This is a flowchart of forest logging detection based on deep learning algorithms provided in an embodiment of this application;
[0016] Figure 3 This is a structural diagram of the forest logging detection device provided in the embodiments of this application;
[0017] Figure 4 This is a structural diagram of the electronic device provided in the embodiments of this application. Detailed Implementation
[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0019] See Figure 1 , Figure 1 This is a flowchart of the forest logging detection method provided in the embodiments of this application. For example... Figure 1 As shown, this embodiment provides a method for detecting deforestation in forest areas, including the following steps:
[0020] Step 101: If the first feature is detected in the first forest area image, detect whether the second feature is detected in the second forest area image. The first forest area image is acquired by a detection device installed on an outdoor tower. The second forest area image includes all or part of the features of the first forest area image. The second forest area image includes the first feature. The first feature is a feature that matches the feature of the logging equipment. The second feature is a feature that matches the feature of the tree logging.
[0021] Step 102: If the second feature exists in the second forest area image, determine that there is logging activity in the forest area corresponding to the second forest area image.
[0022] For ease of understanding and explanation, the following description uses a forest logging detection device as the implementing entity to illustrate the forest logging detection method of this application.
[0023] Before step 101, the forest logging detection device can acquire a first forest area image collected by the detection equipment and detect whether the first forest area image contains a first feature. The detection equipment can be a high-magnification camera, which can be installed on an outdoor tower such as a signal tower or wind power equipment. The first feature is the logging equipment feature, which can be understood as the equipment used to cut down trees, such as logging trucks, trucks, and chainsaws. Correspondingly, the logging equipment feature can be the logging truck feature, truck feature, chainsaw feature, etc.
[0024] If the first feature is detected in the first forest area image, it indicates that logging equipment may have entered the forest area corresponding to the first forest area image, and logging may have occurred in that forest area. At this time, the forest logging detection device cannot determine whether logging has actually occurred in the forest area. For ease of understanding, detecting the first feature in the first forest area image can be interpreted as a suspicious target entering the forest area, which can be understood as suspected logging activity occurring in the forest area.
[0025] If logging has occurred in a forest area, the ground will typically contain fallen tree trunks, sawn tree cuts, and scattered branches. Based on this, in step 101, if the first forest area image detects the presence of a first feature, the forest logging detection device can further detect whether a second feature exists in a second forest area image, thereby determining whether logging has occurred in the forest area corresponding to the second forest area image. The second feature is a tree logging characteristic, such as fallen tree trunks, sawn tree cuts, or scattered branches.
[0026] If a second feature is detected in the second forest area image, it indicates that logging has occurred in that forest area. Based on this, in step 102, when a second feature is detected in the second forest area image, the forest logging detection device can determine that logging has occurred in the forest area corresponding to the second forest area image.
[0027] It should be noted that the first and second forest area images can be images collected by the detection equipment at the same time or at different times. If they are images collected by the detection equipment at the same time, the second forest area image can be an image that is exactly the same as the first forest area image, or it can be an image obtained by magnifying the first forest area image.
[0028] In this embodiment, a detection device installed on an outdoor tower collects images of forest areas. If features matching logging equipment are detected in the forest area image, further detection is performed to determine if features matching tree felling are present. This allows for the determination of whether logging activity has occurred in the corresponding forest area. In this embodiment, since outdoor towers are typically located in high-altitude forestry areas, the detection device installed on the tower can detect the area from top to bottom, covering a large detection area and enabling wide-area forest area detection at a low cost. Furthermore, feature detection of the forest area images allows for timely detection of logging activities, improving the real-time performance of logging detection.
[0029] In some alternative implementations, the first forest area image is acquired by a detection device installed on a signal tower.
[0030] Currently, signal towers are becoming increasingly widespread, and they are generally located in high-altitude areas such as forest areas. The height of the signal towers is usually quite high, and the detection equipment installed on the signal towers can detect the forest area from top to bottom, providing a good detection field of view and providing a good foundation for logging detection in forest areas.
[0031] In this embodiment of the application, the above-mentioned forest logging detection process can be implemented based on deep learning algorithms.
[0032] In some alternative implementations, prior to the step of detecting whether the second forest area image contains a second feature, the method further includes:
[0033] Based on the SSD algorithm, the presence of the first feature in the first forest area image is detected.
[0034] The Single Shot MultiBox Detector (SSD) algorithm is an object detection algorithm. Each cell in the SSD algorithm has a prior box with a different scale or aspect ratio. The predicted bounding boxes are based on these prior boxes, which can reduce the training difficulty to some extent. For example, each cell in the SSD algorithm can have four prior boxes with different scales and aspect ratios; during training, the prior boxes that best suit their shapes can be used.
[0035] In this implementation, the SSD algorithm can be used to detect suspicious targets.
[0036] As an example, firstly, the forest logging detection device can identify the forest area through the SSD algorithm to obtain the specific forest area range (i.e., the forest area range corresponding to the first forest area image); secondly, the forest logging detection device can perform target detection and semantic segmentation on the first forest area image through the SSD algorithm to detect whether there are suspicious target features (i.e., the first feature) in the first forest area image. This process can be understood as performing suspicious target detection or suspected logging behavior detection on the forest area range.
[0037] In the above process, the SSD algorithm outputs a set of detection results (including a bounding box and a confidence score) for each detection box of each unit. Furthermore, the SSD algorithm can treat the background as a special target category. Assuming there are N suspected target categories, the SSD algorithm will predict N+1 detection results, one of which (such as the first detection result) is used to indicate a background target. This implementation is applicable to outdoor environments where the target size is unpredictable and may fill the entire detection image.
[0038] When suspicious target features are present in the first forest area image, the forest logging detection device can assume that there may be suspected logging activity in the forest area and further employ a secondary detection algorithm for depth determination. Depth determination using a secondary detection algorithm can be understood as performing depth detection of tree cuts, fallen trees, and tree stacks based on deep learning algorithms.
[0039] In some alternative implementations, detecting whether the second forest area image contains a second feature includes:
[0040] The second feature is detected in the second forest area image based on the ResNet algorithm.
[0041] In this implementation, the core idea of the ResNet algorithm is to introduce an identity-fast connection structure model, which forms a residual unit of ResNet. The network layers are mainly convolutional layers. For input feature x, the output y of the residual unit in ResNet is:
[0042] y = F(x) + x
[0043] Taking tree cut detection based on the ResNet algorithm as an example, when a tree cut is detected in the target image, it is determined that the tree has been destroyed. The ResNet algorithm has 5 convolutional stages. In the first convolutional stage, the kernel size is 7×7, the number of output feature channels is 64, and the kernel stride is 2. After the first stage of convolution, the size of the image feature is reduced to 0.5 times its original size. Then, in the second convolutional stage, it first goes through a max pooling layer with a pooling window size of 3×3 and a stride of 2. The pooling operation reduces the size of the image feature to 0.5 times its original size. In the subsequent convolutional operations, residual units are stacked. For ResNet with different numbers of layers, the kernel settings of the convolutional layers in the residual units can be different.
[0044] Furthermore, this application embodiment can perform secondary detection based on an improved algorithm of deep learning ResNet101. For the 101-layer ResNet101, a bottleneck design is implemented for the convolutional operations to reduce the number of parameters and computational cost, and a 1×1 convolutional kernel is introduced to perform dimensionality reduction and expansion operations on the number of channels in the feature map. Therefore, the network layer in each residual unit consists of three convolutional layers with kernel sizes of 1×1, 3×3, and 1×1, respectively. Each of the above convolutional layers is connected by a ReLU activation layer. In the conv3_1, conv4_1, and conv5_1 stages, the network performs feature downsampling by setting the convolutional kernel stride to 2. The open-source ResNet101 algorithm model removes the average pooling and fully connected layers used for image classification in the ResNet network, which can effectively improve the detection speed. Furthermore, for occlusions or blurriness in the image, conv1, conv2_x, conv3_x, conv4_x, and conv5_x can be used to extract features from the input features, which can increase the target detection rate and effectively avoid false alarms caused by occlusions or false alarms caused by blurriness.
[0045] See Figure 2 , Figure 2 This is a flowchart illustrating forest logging detection based on a deep learning algorithm, provided in an embodiment of this application. For example... Figure 2 As shown, the process of detecting deforestation in forest areas based on deep learning algorithms may include the following steps:
[0046] Step 201: Input image;
[0047] Step 202: Target detection; Target types can be predefined, including types such as trees and logging equipment. The SSD algorithm can be used to perform initial detection on the image to determine the quantity, location, pixel size, and other information of each type of target in the current image.
[0048] Step 203: Semantic segmentation; The SSD algorithm can be used to fit the target region and range to perform semantic segmentation on the image;
[0049] Step 204: Suspected logging behavior determination; The SSD algorithm can be used to determine whether the detected target is a suspicious target, thereby determining whether there is suspected logging behavior; if so, proceed to step 205, otherwise proceed to step 206;
[0050] Step 205: Secondary detection; The ResNet algorithm can be used to perform secondary detection on the image to determine whether logging has occurred;
[0051] Step 206: Output the detection results.
[0052] The above are the implementation methods for detecting deforestation in forest areas based on deep learning algorithms provided in the embodiments of this application.
[0053] In this embodiment, deforestation detection can be adaptively performed; specifically, deforestation detection can be adaptively performed on forest area images. The following describes the relevant implementation methods for adaptively acquiring forest area images.
[0054] In some alternative implementations, prior to the step of detecting whether the second forest area image contains a second feature, the method further includes:
[0055] The detection device acquires images within its detection range.
[0056] If the detection device is found to have acquired an image of the first forest area, the system detects whether the first feature exists in the image of the first forest area.
[0057] In this embodiment, the detection device can scan at fixed times and locations within its detection range. During the scanning process, the forest logging detection device can identify whether the images collected by the detection device are forest area images. When it identifies an image as a forest area image, the forest logging detection device can perform logging detection. If it does not identify an image as a forest area image, the forest logging detection device does not need to perform logging detection. In other words, the forest logging detection device can adaptively perform logging detection on forest area images without needing to perform logging detection on any images collected by the detection device. This improves detection efficiency and saves resources required for detection.
[0058] In addition, the detection equipment can first acquire images within its detection range, and then the forest logging detection device can perform image recognition from all the acquired images to determine the forest area image, and then perform the logging detection process as described above. This method is relatively easy to understand, so it will not be explained in detail.
[0059] In some alternative implementations, after the step of detecting whether the first feature exists in the first forest area image and before the step of detecting whether the second feature exists in the second forest area image, the method further includes:
[0060] If the first feature is detected in the first forest area image, the second forest area image is acquired by zooming the detection device. The second forest area image includes some features of the first forest area image.
[0061] When the forest logging detection device detects a first feature in the first forest area image, it is necessary to further detect whether a second feature exists in the forest area image. In this embodiment, considering that the detection accuracy of the second feature is higher than that of the first feature, and in order to improve the real-time performance of forest logging detection, the forest logging detection device needs to be able to detect even when the scale of logging in the forest area is small. Therefore, based on the adaptive acquisition of the first forest area image, the second forest area image can also be adaptively acquired by zooming the detection device. In this way, because the forest area image is adaptively magnified, the forest logging detection device can perform logging detection more accurately.
[0062] In this embodiment, the detection device can be a high-magnification variable zoom camera.
[0063] In some optional implementations, if the second feature exists in the second forest area image, determining that there is logging activity in the forest area corresponding to the second forest area image includes:
[0064] If the second feature exists in the second forest area image, calculate the degree of tree felling;
[0065] If the degree of tree felling is greater than or equal to a preset threshold, it is determined that there is logging activity in the forest area corresponding to the second forest area image.
[0066] In this embodiment, considering some factors, the presence of a very small number of fallen trees or tree cuts does not necessarily indicate that logging has occurred in the forest area. In order to avoid false detections caused by accidental factors and improve the accuracy of logging detection, when the second forest area image has a second feature, the forest area logging detection device can calculate the degree of tree felling. If the degree of tree felling is greater than or equal to a preset threshold, it is determined that logging has occurred in the forest area corresponding to the second forest area image.
[0067] The degree of deforestation can be calculated based on the number, coverage, or accumulation of the second feature in the second forest area image.
[0068] In summary, the beneficial effects of this application include at least the following: by collecting forest area images through detection equipment installed on an outdoor tower and performing feature detection on the forest area images, the occurrence of logging activities can be detected in a timely manner, improving the real-time performance of logging detection, and covering a large detection area, thereby achieving large-scale forest area detection at a low cost.
[0069] See Figure 3 , Figure 3 This is a structural diagram of the forest logging detection device provided in the embodiments of this application, as shown below. Figure 3 As shown, the forest logging detection device 300 includes:
[0070] The first detection module 301 is used to detect whether a second feature exists in a second forest area image when a first feature is detected in a first forest area image. The first forest area image is acquired by a detection device installed on an outdoor tower. The second forest area image includes all or part of the features of the first forest area image. The second forest area image includes the first feature. The first feature is a feature that matches the feature of the logging equipment. The second feature is a feature that matches the feature of the tree logging.
[0071] The determining module 302 is used to determine that there is logging activity in the forest area corresponding to the second forest area image when the second feature exists in the second forest area image.
[0072] In some embodiments, the forest logging detection device 300 further includes:
[0073] The first acquisition module is used to acquire images within the detection range of the detection device.
[0074] The second detection module is used to detect whether the first feature exists in the first forest area image when the detection device has collected the first forest area image.
[0075] In some embodiments, the forest logging detection device 300 further includes:
[0076] The second acquisition module is used to acquire a second forest area image by zooming the detection device when the first forest area image is detected to contain the first feature. The second forest area image includes some features of the first forest area image.
[0077] In some embodiments, the determining module 302 includes:
[0078] The calculation unit is used to calculate the degree of tree felling when the second feature exists in the second forest area image;
[0079] The determining unit is used to determine that there is logging activity in the forest area corresponding to the second forest area image when the degree of tree felling is greater than or equal to a preset threshold.
[0080] In some embodiments, the forest logging detection device 300 further includes:
[0081] The third detection module is used to detect whether the first feature exists in the first forest area image based on the SSD algorithm.
[0082] In some embodiments, the first detection module 301 is specifically used for:
[0083] Based on the ResNet algorithm, the presence of the second feature in the second forest area image is detected.
[0084] In some embodiments, the first forest area image is acquired by a detection device installed on a signal tower.
[0085] The 300-ton forest logging detection device can achieve... Figures 1 to 2 The various processes implemented in the illustrated method embodiments achieve the same beneficial effects, and will not be described again here to avoid repetition.
[0086] According to embodiments of this application, this application also provides an electronic device and a readable storage medium.
[0087] like Figure 4 The diagram shown is a block diagram of an electronic device according to an embodiment of this application. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.
[0088] like Figure 4As shown, the electronic device includes one or more processors 401, a memory 402, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components are interconnected via different buses and can be mounted on a common motherboard or otherwise as required. The processors can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In other embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple electronic devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 401 as an example.
[0089] The memory 402 is the computer-readable storage medium provided in this application. The memory stores instructions executable by at least one processor to cause the at least one processor to perform the forest logging detection method provided in this application. The computer-readable storage medium of this application stores computer instructions for causing a computer to perform the forest logging detection method provided in this application.
[0090] Memory 402, as a computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the forest logging detection method in the embodiments of this application (e.g., attached...). Figure 3 The first detection module 301 and the determination module 302 are shown. The processor 401 executes various functional applications and data processing of the server by running non-transient software programs, instructions and modules stored in the memory 402, thereby realizing the forest logging detection method in the above method embodiment.
[0091] The memory 402 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device implementing the forest logging detection method. Furthermore, the memory 402 may include high-speed random access memory and may also include non-transient memory, such as at least one disk storage device, flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 402 may optionally include memory remotely located relative to the processor 401, and these remote memories can be connected to the electronic device implementing the forest logging detection method via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0092] The electronic device for implementing the forest logging detection method may further include: an input device 403 and an output device 404. The processor 401, memory 402, input device 403, and output device 404 can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.
[0093] Input device 403 can receive input digital or character information, as well as key signal input related to user settings and function control of electronic equipment implementing the forest logging detection method, such as touch screen, keypad, mouse, trackpad, touchpad, indicator, one or more mouse buttons, trackball, joystick, etc. Output device 404 may include display device, auxiliary lighting device (e.g., LED), and haptic feedback device (e.g., vibration motor). The display device may include, but is not limited to, liquid crystal display (LCD), light-emitting diode (LED) display, and plasma display. In some embodiments, the display device may be a touch screen.
[0094] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, application-specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.
[0095] These computational programs (also referred to as programs, software, software applications, or code) include machine instructions for a programmable processor and can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms “machine-readable medium” and “computer-readable medium” refer to any computer program product, device, and / or apparatus (e.g., disk, optical disk, memory, programmable logic device (PLD)) used to provide machine instructions and / or data to a programmable processor, including machine-readable media that receive machine instructions as machine-readable signals. The term “machine-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor.
[0096] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).
[0097] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.
[0098] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0099] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.
Claims
1. A method for detecting logging in forest areas, characterized in that, include: When the SSD algorithm detects the presence of a first feature in the first forest area image, the detection device zooms in to acquire a second forest area image and detects whether the second forest area image has a second feature. The first forest area image is acquired by the detection device installed on the outdoor tower. The second forest area image includes some features of the first forest area image. The second forest area image is a magnified partial image of the first forest area image. The second forest area image includes the first feature, which is a feature that matches the features of the logging equipment. The second feature is a feature that matches the features of the tree felling. The detection of whether the second forest area image has a second feature includes: performing depth detection of tree cuts, fallen trees, and piled trees in the second forest area image based on a residual neural network algorithm; If the second feature exists in the second forest area image, the tree felling degree is calculated, and if the tree felling degree is greater than or equal to a preset threshold, it is determined that there is felling behavior in the forest area corresponding to the second forest area image; wherein, the tree felling degree is calculated based on the number, coverage or accumulation of the second feature in the second forest area image.
2. The method according to claim 1, characterized in that, Prior to the step of detecting whether the second forest area image contains a second feature, the method further includes: The detection device acquires images within its detection range. If the detection device is found to have acquired an image of the first forest area, the system detects whether the first feature exists in the image of the first forest area.
3. The method according to any one of claims 1 to 2, characterized in that, Prior to the step of detecting whether the second forest area image contains a second feature, the method further includes: Based on the SSD algorithm, the presence of the first feature in the first forest area image is detected.
4. The method according to any one of claims 1 to 2, characterized in that, The detection of whether the second forest area image contains a second feature includes: Based on the ResNet algorithm, the presence of the second feature in the second forest area image is detected.
5. The method according to any one of claims 1 to 2, characterized in that, The first forest area image was acquired using detection equipment installed on a signal tower.
6. A forest logging detection device, characterized in that, include: The first detection module is used to detect whether a second feature exists in the second forest area image when a first feature is detected in the first forest area image based on the SSD algorithm. The second forest area image is acquired by a detection device that is zoomed in on the first forest area image. The second forest area image includes some features of the first forest area image. The second forest area image is a magnified partial image of the first forest area image. The second forest area image includes the first feature, which is a feature that matches the feature of the logging equipment. The second feature is a feature that matches the feature of the tree felling. The detection of whether the second forest area image has a second feature includes: performing depth detection of tree cuts, fallen trees, and piled trees in the second forest area image based on a residual neural network algorithm; The determination module is used to calculate the tree felling degree when the second feature exists in the second forest area image, and to determine that there is logging behavior in the forest area corresponding to the second forest area image when the tree felling degree is greater than or equal to a preset threshold; wherein the tree felling degree is calculated based on the number, coverage or accumulation of the second feature in the second forest area image.
7. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the forest logging detection method as described in any one of claims 1 to 5.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the forest logging detection method as described in any one of claims 1 to 5.