KR-20260066222-A - System for analyzing carbon absorption of forest and method thereof

KR20260066222AKR 20260066222 AKR20260066222 AKR 20260066222AKR-20260066222-A

Abstract

The present invention relates to a method for analyzing forest carbon absorption, comprising: a first step of collecting image data captured by a drone in a forest area through a collection unit; a second step of converting the collected image data into a 3D point cloud using a NeRF model through a generation unit; a third step of analyzing individual tree data within the forest in the 3D point cloud through a processing unit; a fourth step of inputting the individual tree clusters into a PointNet model to classify tree species through a classification unit; a fifth step of calculating the total carbon absorption of the forest through an analysis unit based on the tree species and count data; and a sixth step of displaying the carbon absorption amount and tree species distribution on a screen through an output unit or storing them in a database.

Inventors

장민석

Assignees

장민석

Dates

Publication Date: 20260512
Application Date: 20241104

Claims (4)

Step 1: Collecting image data of a forest area captured by a drone through a collection unit; A second step of converting the collected image data into a 3D point cloud using a NeRF model through a generation unit; A third step of analyzing individual tree data within the forest from the above 3D point cloud through a processing unit; A fourth step of inputting the above individual tree clusters into a PointNet model to classify tree species through a classification unit; A fifth step of calculating the total carbon absorption amount of the forest through an analysis unit based on the above tree species and number data; and A method for analyzing forest carbon absorption, characterized by including: a sixth step of displaying the carbon absorption amount and tree species distribution on a screen through an output unit or storing them in a database.
In paragraph 1, The above third step is a method for analyzing forest carbon absorption, characterized by separating forests and terrain from a point cloud generated through a Segmentation model to retain only necessary forest data and removing terrain or unnecessary elements, and clustering individual trees within a forest area through a Clustering model to identify each tree as a single cluster and calculate the number of trees.
A collection unit that collects image data by photographing a forest area using a high-resolution camera mounted on a drone; A generation unit that converts the above-mentioned collected image data into a 3D point cloud using a NeRF model; A processing unit that analyzes individual tree data within the forest in the above 3D point cloud; A classification unit that classifies tree species by inputting the above individual tree clusters into a PointNet model; An analysis unit that calculates the total carbon absorption of the forest based on the data of the species and number of the trees mentioned above; and A forest carbon absorption analysis system characterized by including an output unit that displays the carbon absorption amount and tree species distribution on a screen or stores them in a database.
In paragraph 3, A forest carbon absorption analysis system characterized by the above-mentioned processing unit separating forests and terrain from a point cloud generated through a Segmentation model, removing terrain or unnecessary elements while leaving only necessary forest data, clustering individual trees within a forest area through a Clustering model to identify each tree as a single cluster, and calculating the number of trees.

Description

System for analyzing carbon absorption of forest and method thereof The present invention relates to a forest carbon absorption analysis system and method, and more specifically, to a forest carbon absorption analysis system and method that calculates the carbon absorption amount of a forest captured by a drone using artificial intelligence. In modern society, climate change caused by carbon emissions is creating serious social and environmental problems. To address these issues, interest in carbon neutrality and carbon absorption is growing. In particular, as forests are major carbon sinks, it is crucial to accurately measure and analyze their carbon absorption capacity. However, conventional methods for measuring carbon absorption primarily rely on field surveys and simulations; however, these methods are time-consuming and costly and have limitations in providing precise data, necessitating the need for new measurement technologies to enhance accuracy. Figure 1 is a block diagram showing a forest carbon absorption analysis system according to the present invention. Figure 2 is a flowchart illustrating a method for analyzing forest carbon absorption according to the present invention. Embodiments of the present invention are described below with reference to the attached drawings so that those skilled in the art can easily implement the invention. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Now, a preferred embodiment of the forest carbon absorption analysis system and method according to the present invention will be described in detail. Figure 1 is a block diagram showing a forest carbon absorption analysis system according to the present invention. Referring to FIG. 1, the forest carbon absorption analysis system (1000) according to the present invention may include a collection unit (100), a generation unit (200), a processing unit (300), a classification unit (400), an analysis unit (500), and an output unit (600). The above collection unit (100) can collect image data by photographing the forest area from various angles using a high-resolution camera mounted on a drone. The generation unit (200) can convert collected image data into a 3D point cloud using a NeRF (Neural Radiance Fields) model. Here, the NeRF can generate 3D data by reconstructing a 2D image into 3D. The above processing unit (300) can precisely analyze individual tree data within a forest in a 3D point cloud. The above processing unit (300) can separate forests and terrain from a point cloud generated through a Segmentation model, and improve the analysis accuracy by removing terrain or unnecessary elements while leaving only the necessary forest data, and can cluster individual trees within a forest area through a Clustering model to identify each tree as a single cluster and automatically calculate the number of trees. The above classification unit (400) can classify tree species by inputting individual tree clusters into a PointNet model. Through this, the distribution of various tree species within the forest can be identified. The above analysis unit (500) can calculate the total carbon absorption of the forest based on tree species and tree count data according to the Korea Forest Service formula (total carbon absorption (tC) = biomass × carbon conversion factor × carbon content). Through this, the contribution of the forest to climate change response can be evaluated. The output unit (600) can display the final analysis results, such as carbon absorption amount and tree species distribution, on a screen or store them in a database for use. This data can be utilized for forest management and carbon offset schemes. Figure 2 is a flowchart illustrating a method for analyzing forest carbon absorption according to the present invention. Referring to FIG. 2, the method for analyzing forest carbon absorption according to the present invention first collects image data of a forest area captured through a collection unit (100) (S100). Here, the forest area may be captured using a drone equipped with a high-resolution camera, but is not limited thereto. Next, the collected image data is converted into a 3D point cloud using a NeRF (Neural Radiance Fields) model through a generation unit (200) (S200). Next, individual tree data within the forest in the generated point cloud is analyzed precisely through the processing unit (300) (S300). Here, the processing unit (300) separates the forest and terrain from the point cloud generated through the Segmentation model, removes terrain or unnecessary elements while leaving only the necessary forest data to improve the analysis accuracy, clusters individual trees within the forest area through the Clustering model to identify each tree as a single cluster, and automatically calculates the number of trees. Next, the individual tree clusters are input into a PointNet model to classify the tree species through a c