通讯机构：
[Tan, SQ ] C;Cent South Univ Forestry & Technol, Coll Forestry, Changsha 410004, Peoples R China.
关键词：
forest fire scenes;visible and infrared images;image fusion;deep learning;GAN
摘要：
<jats:p>Aimed at addressing deficiencies in existing image fusion methods, this paper proposed a multi-level and multi-classification generative adversarial network (GAN)-based method (MMGAN) for fusing visible and infrared images of forest fire scenes (the surroundings of firefighters), which solves the problem that GANs tend to ignore visible contrast ratio information and detailed infrared texture information. The study was based on real-time visible and infrared image data acquired by visible and infrared binocular cameras on forest firefighters’ helmets. We improved the GAN by, on the one hand, splitting the input channels of the generator into gradient and contrast ratio paths, increasing the depth of convolutional layers, and improving the extraction capability of shallow networks. On the other hand, we designed a discriminator using a multi-classification constraint structure and trained it against the generator in a continuous and adversarial manner to supervise the generator, generating better-quality fused images. Our results indicated that compared to mainstream infrared and visible image fusion methods, including anisotropic diffusion fusion (ADF), guided filtering fusion (GFF), convolutional neural networks (CNN), FusionGAN, and dual-discriminator conditional GAN (DDcGAN), the MMGAN model was overall optimal and had the best visual effect when applied to image fusions of forest fire surroundings. Five of the six objective metrics were optimal, and one ranked second-to-optimal. The image fusion speed was more than five times faster than that of the other methods. The MMGAN model significantly improved the quality of fused images of forest fire scenes, preserved the contrast ratio information of visible images and the detailed texture information of infrared images of forest fire scenes, and could accurately reflect information on forest fire scene surroundings.</jats:p>
关键词：
forest fire smoke;Detection model;Convolutional Neural Network;vision Transformer;Lightweight model
摘要：
Forest fires seriously jeopardize forestry resources and endanger people and property. The efficient identification of forest fire smoke, generated from inadequate combustion during the early stage of forest fires, is important for the rapid detection of early forest fires. By combining the Convolutional Neural Network (CNN) and the Lightweight Vision Transformer (Lightweight ViT), this paper proposes a novel forest fire smoke detection model: the SR-Net model that recognizes forest fire smoke from inadequate combustion with satellite remote sensing images. We collect 4,000 satellite remote sensing images, 2,000 each for clouds and forest fire smoke, from Himawari-8 satellite imagery located in forest areas of China and Australia, and the image data are used for training, testing, and validation of the model at a ratio of 3:1:1. Compared with existing models, the proposed SR-Net dominates in recognition accuracy (96.9%), strongly supporting its superiority over benchmark models: MobileNet (92.0%), GoogLeNet (92.0%), ResNet50 (84.0%), and AlexNet (76.0%). Model comparison results confirm the accuracy, computational efficiency, and generality of the SR-Net model in detecting forest fire smoke with high temporal resolution remote sensing images.
作者机构：
[Zeng, Siqi; Yang, Shengyang; Zeng, SQ; Long, Shisheng; Xiao, Huashun] Cent South Univ Forestry & Technol, Fac Forestry, Changsha 410004, Peoples R China.;[Gong, Zhaosong] Sichuan Forestry Survey & Planning Inst, Chengdu 610081, Peoples R China.
通讯机构：
[Zeng, SQ ] C;Cent South Univ Forestry & Technol, Fac Forestry, Changsha 410004, Peoples R China.
关键词：
maximum size-density;stand parameters;growth balance status;isogonic growth;growth rate
摘要：
Stand density management is important for decision-making regarding adaptive silviculture and thinning, growth modelling, and yield prediction in forests, especially plantations. Although substantial research related to the self-thinning rule and maximum size-density law has been conducted, there are still critical gaps that exist in the biophysical explanation and validation of the relationships among stand variables and relevant parameters. In this study, time series observations from six plots of fully stocked Chinese fir plantations with different densities of planted trees were used to characterise the growth of stand basal area (G), average height (H), and diameter at breast height (D). The growth trends in the stand parameters and the relationships among them were analysed. As indicated by previous studies, in the fully stocked stands, there was a significant linear relationship between G and H. This study also resulted in the following new findings: (1) At the beginning, the growth rate of stand basal area (PG) was greater than the growth rate of average height (PH), but PG decreased quickly as the stands approached canopy closure and then became stable. Meanwhile, as the stands neared canopy closure, the rate of increase in the G/H ratio decelerated, ultimately resulting in a stable G/H value that approached the first limit value. This led to a stand growth balance status that continued until self-thinning took place. (2) Artificial thinning broke the growth balance status, but the stands returned to balance status if they were still young enough. Self-thinning also broke the growth balance status and lead to fluctuating growth rates of both G and H, but the fluctuations were very slight, which showed a trend in similar growth rates of G and H. (3) The findings implied that the stand G and H growths were allometric at the beginning but became isogonic as canopy closure and self-thinning were approached. On the other hand, the H growth rate was generally greater than that of D, but both growth rates showed a trend in similar values after the stands matured. Subsequently, the H/D ratio is anticipated to stabilize and gradually converge towards the second limit value once the stands reach maturity. The results implied that the stand growth balance status and two limit values can be used to identify and select fully stocked stands that are needed for the development of maximum size-density equations and self-thinning rules.
