69 results about "Mountain forest" patented technology

The invention relates to the technical field of intelligent path planning, in particular to an intelligent path planning method and system for hilly and mountain forest machine operation based on satellite remote sensing, and is suitable for intelligent path planning of forestry mechanized operation under complex terrain conditions. Building a multi-dimensional energy consumption prediction model in combination with forest machine parameter data; an initial path is generated by adopting layered progressive path planning, and dynamic adjustment is performed by utilizing real-time environment data so as to adapt to a complex mountain environment; under the scene of multi-machine collaborative operation, the efficiency and safety of the final operation path are ensured through optimization of the adjusted path, the method accurately recognizes the dangerous area, optimizes the passing path, remarkably improves the safety and efficiency of forest machine operation, and is suitable for the complex operation environment of hills and mountains.

The invention discloses a lightweight small target detection method based on improved YOLOv11, and relates to the technical field of artificial intelligence and computer vision. Comprising the following steps: acquiring a mountain forest region image containing a small target, and performing data preprocessing on the mountain forest region image containing the small target to obtain a small target data set; a C3K2 module of the YOLOv11 model is replaced with a CGBlock module, two initial conv modules at the input end of a backbone network are reserved, the other conv modules are replaced with WP modules, an original detection head is replaced with an LDHead detection head, and the improved YOLOv11 model is constructed; inputting a small target data set, continuously iterating to obtain an optimal parameter, and generating a trained improved YOLOv11 model; and small target objects in the mountain forest area are detected. The method has the advantages of being light in structure, high in detection precision and high in complex background adaptability, and is suitable for small target detection tasks in a resource-constrained environment.

The invention provides a fire-fighting virtual prevention simulation system and method for a mountain forest region, and the method comprises the steps: extracting a terrain elevation change trend according to a mountain numerical model, and calculating a spreading path of a fire behavior under different terrain conditions in combination with vegetation density distribution; a meteorological data real-time updating module is adopted to obtain current meteorological conditions and humidity data, and prediction parameters of the fire spreading speed are dynamically adjusted; if the terrain elevation change exceeds a preset threshold value, calculating an acceleration effect of the fire behavior on the steep slope, and correcting a spreading path prediction result; if the vegetation density distribution shows valley characteristics, the chimney effect of the fire behavior in valleys is analyzed, and the fire development situation prediction map is updated; according to the soil type and the geological structure, combustion characteristics of different areas are judged, and a calculation formula of the fire spreading speed is adjusted; and adopting a machine learning algorithm, fusing the terrain elevation, the vegetation density and the meteorological condition data, and generating a prediction result of the fire spreading path.

The invention relates to the technical field of remote sensing biomass detection, in particular to a mountain forest biomass recognition method and system based on high-resolution remote sensing data, and the method comprises the steps: dividing all clusters into a high-gradient shady slope cluster and a low-gradient sunny slope cluster based on the gradient feature value of each pixel; based on the pixel values of all the pixels in each low-gradient sunny slope cluster, screening out feature pixels from each low-gradient sunny slope cluster; and based on a result of fusion of a mean value of pixel values of all pixels in each high-gradient shady slope cluster and a texture feature value, and in combination with a confusion degree of coordinates of all feature pixels in each low-gradient sunny slope cluster, determining biological complexity, and identifying the biomass of the mountain forest region to be identified. According to the method, the problem that the biomass in the mountain forest is misestimated due to interference of topographic factors is solved, and the accuracy of identifying the biomass condition of the mountain forest is improved.

The invention relates to the technical field of fire source positioning path planning, in particular to an unmanned aerial vehicle fire source point positioning path planning method based on machine learning. Comprising the following steps: firstly, acquiring a suspected fire source estimated position of a mountain forest, and establishing a fire source attraction vector field according to position information and environment information of a suspected fire source area; secondly, generating an accurate initial fire source search path according to influence information in the fire source vector field; in the searching process, a fire source point recognition model is adopted, the model carries out learning from multi-dimensional features according to image information and numerical information, the accuracy of fire source recognition is improved, and the position of a fire source can be adjusted according to a recognition result; and finally, establishing an obstacle repulsion field by using obstacle information of the mountain forest, and combining the obstacle repulsion field with the fire source attraction vector field to form a total influence vector field optimization search path. According to the method, the accuracy of unmanned aerial vehicle fire source point positioning path planning is improved.

The invention discloses a fabricated integrated house (SWP), and belongs to the technical field of fabricated buildings. Comprising a foundation part 01 which is a cast-in-place concrete foundation or a spiral pile foundation and the like; the ground 02 comprises floor bearing plates and the like; 03, a steel structure part; the outer wall comprises an outer facing layer of the outer wall, a wall body of the outer wall and an inner facing layer of the outer wall, or the outer wall is a mortise and tenon structure wall body; doors and windows 08, indoor suspended ceilings 09, metal tiles and the like; the system further comprises an inner partition wall 07 and a roof heat preservation and insulation layer 10. The system can also be of a two-storey or more structure. The invention provides an integral house solution for people who are deficient in concrete or inconvenient to transport in national grassland areas or mountain forest areas of Southeast Asia; the structure is stable, fixed and compact; and factory manufacturing, field installation and construction are convenient, the speed is high, the cost is low, and the cost performance is high. The screw pile foundation is suitable for soil land; the concrete foundation is wide in application range; and the wall body is hollow and is heat-insulated. The system can also be used for building communities in batches.

The invention discloses a steep mountainous area forest height inversion method and device fused with rising and falling rail short wave InSAR data, and aims to solve the problems of insufficient observation information and steep mountainous area gradient decoherence during InSAR inversion, and the method comprises the steps: obtaining rising and falling rail short wave InSAR data and sparse LiDAR data, and carrying out the preprocessing; calculating the gradient by using a digital elevation model, and performing terrain effect correction on the effective vertical wave number; determining a lifting rail data overlapping region, and establishing a semi-empirical SINC model for a non-overlapping region to invert the forest height; constructing a double-baseline forest height inversion model considering coherent phase and amplitude at the same time for the overlapped region, determining an optimal model parameter by taking the sparse LiDAR forest height as control, and performing inversion on the forest height; and finally, inlaying forest heights of the overlapping region and the non-overlapping region to form a complete and continuous product. According to the method, high-precision forest height inversion of the steep mountainous area is realized by fusing the ascending and descending rail InSAR data, and the method has good stability and popularization value.

The invention discloses a mountain forest global intelligent fire monitoring and automatic disposal system. The system comprises a rainwater collection and water polo fire extinguishing unit, a wind-solar complementary power supply and robot dog battery replacement unit, a multi-sensor monitoring unit, an unmanned aerial vehicle disposal unit and an intelligent control center. The system is based on theoretical design and engineering planning, rainwater is collected and filled into a specially-made rubber water ball, and accurate throwing fire extinguishing of the unmanned aerial vehicle is achieved; wind-solar power generation and robot dog autonomous power conversion design is adopted, so that the power supply continuity of the system is ensured; through the multi-level multi-sensor fusion design, low false alarm and rapid identification of the fire behavior are realized; and a three-level fire disposal strategy is designed, and automatic disposal and manual remote takeover are supported. The water storage tank is of a heat preservation heating structure, and annual stable operation is achieved. The system is suitable for fire prevention of large-area mountain forests, realizes water source self-sufficiency, autonomous power supply, intelligent monitoring, automatic fire extinguishing and autonomous operation and maintenance in principle, and remarkably improves the early-stage handling capacity of mountain forest fires.

The invention provides a method for correcting satellite laser footprint positioning errors in mountainous forest areas, comprising: acquiring laser echo waveform data of the mountainous forest area, calculating the true laser ranging value of the mountainous forest canopy of each laser footprint; selecting points near the laser footprint as possible laser footprint points, calculating the satellite-ground integrated laser ranging value of the possible laser footprint points, and the ranging residuals between the satellite-ground integrated laser ranging value and the true laser ranging value of the mountainous forest canopy; traversing a number of continuous laser footprints, calculating the ranging residuals of the possible laser footprint points and accumulating them to obtain the ranging residual sum; iteratively calculating the ranging residual sum of the possible laser footprint points, and taking the possible laser footprint points corresponding to the minimum ranging residual sum as correction points, so as to complete the correction of the laser footprint positioning errors in the mountainous forest area without using high-precision terrain data.

The invention relates to the technical field of biodiversity research, and discloses a mountain tree species diversity remote sensing estimation method, which comprises the following steps: acquiring mountain tree species distribution data, acquiring a mountain remote sensing image, preprocessing the data, and calculating a tree species diversity index according to tree species distribution. Calculating the correlation between the spectral heterogeneity index and the tree species diversity index, constructing a mountain tree species diversity remote sensing mapping model according to the influence factor calculated by the spectral heterogeneity index, extracting the optimal parameter of mountain tree species diversity remote sensing estimation, and making a tree species diversity index spatial distribution map according to the spectral heterogeneity index to realize tree species diversity estimation. A terrain correction mechanism is introduced into the spectral heterogeneity index constructed by the method, the stability and reliability of the spectral heterogeneity index are improved, spectral response of the model to different tree species is improved after a weighted distance construction mechanism between pixels is combined, and the method is suitable for mountain forest areas with obvious topographic relief and has good application prospects. And multi-source input of multi-sensor remote sensing images is supported.

The invention discloses an image data acquisition method and device based on an unattended intelligent airport, and relates to the technical field of image data processing.The method comprises the steps that forest image data are acquired based on the unattended intelligent airport, the unattended intelligent airport selects a forest area meeting preset mountain forest conditions, and the selected forest area is selected as a forest area; the unattended intelligent airport is used for taking off, landing and parking a predetermined type of unmanned aerial vehicle, and the predetermined mountain forest conditions comprise that the height fall of a mountain land is greater than a preset fall threshold value, the forest vegetation coverage rate is higher than a preset coverage rate threshold value, and the number of roads in a forest is lower than a preset number threshold value; and analyzing the preprocessed forest image data by adopting a pre-trained machine learning model, identifying a forest resource state, and displaying the forest resource state on a terminal screen. The technical problems that image resource analysis is difficult to carry out on an unattended intelligent airport system in a severe environment and the operation efficiency is low in the prior art are solved.

1. Name of the designed product: sofa (mountain forest). 2. Use of the designed product: sofa for sitting when resting. 3. Design points of the designed product: in shape. 4. Picture or photo best indicating the design points: perspective view.

The application relates to the technical field of remote sensing biomass detection, in particular to a mountain forest biomass recognition method and system based on high-resolution remote sensing data.The method comprises the following steps: based on the slope characteristic value of each pixel, all clustering clusters are divided into high-slope shady slope clusters and low-slope sunny slope clusters; based on the pixel value of all pixels in each low-slope sunny slope cluster, feature pixels are screened out from each low-slope sunny slope cluster; based on the fusion result of the pixel value mean value and the texture characteristic value of all pixels in each high-slope shady slope cluster, and in combination with the confusion degree of the coordinates of all feature pixels in each low-slope sunny slope cluster, the biological complexity is determined, and the biomass of a to-be-recognized mountain forest region is recognized. The application solves the problem that the biomass in the mountain forest is wrongly estimated due to the interference of terrain factors, and improves the accuracy of the recognition of the biomass condition of the mountain forest.

The invention discloses a rare earth mining area illegal mining unmanned aerial vehicle remote sensing intelligent detection method fusing multi-scale convolution and an attention mechanism. According to the method, YOLO11m is used as a baseline network framework, a multi-scale convolution attention enhancement module is embedded in the feature connection position of a backbone network and a neck network of the YOLO11m, and the multi-scale convolution attention enhancement module is formed by cascading a channel attention module, a space attention module and a multi-scale convolution residual block module. Complementary collaboration is formed among semantic channel enhancement, spatial region focusing and multi-scale context modeling through a progressive enhancement link, so that the detection rate and the positioning stability of a suspected illegal mining target under the complex mountain forest region condition are improved.

The application relates to the technical field of environmental protection, in particular to a forest desertification treatment method, which comprises the following steps: firstly, preparing a planting layer; then, transporting the planting layer to a desertification area; then, filling nutrient soil in the planting layer; then, planting plants in the nutrient soil; then, laying the planting layer on the surface of a stone wall; then, fixing the planting layer; then, watering the surface of the planting layer; and finally, spraying sticky mud on the surface of the planting layer. The planting layer is laid and fixed on the surface of the stone wall, and plants are planted in planting cavities in the planting layer, so that the influence of the slope gradient of the stone wall on the planting of the plants can be ignored, green plants can be effectively covered on steep stone walls, especially on huge stone blocks forming cliffs in mountain forests, and the desertification treatment effect of the mountain forests is improved.

The application discloses a forest leaf area index inversion acquisition method and system based on data fusion. The acquisition method comprises the following steps: performing cloud removal processing and cutting on remote sensing image data of a research area to obtain a segmented image; based on land cover data, obtaining forest pixels in each segmented image, and performing spatial aggregation matching with resolution to obtain aggregated remote sensing images; extracting a vegetation index from the segmented image; obtaining forest canopy height data of the research area; taking reflectivity values of different bands in the aggregated remote sensing images, the vegetation index, the forest canopy height data and a sun sensor related parameter as inputs, and taking the forest leaf area index as output to construct a forest leaf area index fusion prediction model, and obtaining a prediction result. The LAI distribution result obtained by the application has more spatial details, is more suitable for mountain forest ecosystems with large spatial heterogeneity, and can accurately and rapidly monitor mountain vegetation parameters.

The invention provides a fire monitoring method, which comprises the steps of performing data acquisition on a mountain forest fire scene through data acquisition equipment, performing scanning for flame color distribution and contour shape, and generating an initial fire feature matrix in combination with dynamic features of motion track fluctuation and regional heat accumulation; after the fire risk distribution map is obtained, adjusting key parameters in the fire feature vector by adopting a solution concentration proportioning technology and simulating the influence of a chemical environment stability test on the smoke diffusion rate, and obtaining the optimized fire state description; and according to the dynamic boundary of the fire area, fusing boundary heat flow analysis and solvent evaporation rate logic, analyzing the change trend of the smoke concentration gradient in the fire area, and generating the fire diffusion prediction sequence.

The present application relates to the technical field of mountain forest fertilization, and discloses a mountain forest tree energy-saving and environment-friendly slow-release fertilizer device, which comprises an enclosing cylinder, the enclosing cylinder is provided with an enclosing cavity, the enclosing cavity is provided with a swing ring plate, the swing ring plate is provided with an outer circumferential ring and an inner circumferential ring, the inner circumferential ring surrounds a tree base and is elastically connected with the tree base, and the outer circumferential ring surrounds the enclosing cavity in a circumferential direction and is elastically connected with the inner side wall of the enclosing cavity; the bottom of the swing ring plate is arranged in a suspended manner, the swing ring plate is provided with a storage cavity for placing fertilizer, and a plurality of fertilizer application holes are arranged in the swing ring plate; the top opening of the enclosing cavity is provided with a floating ring plate, the floating ring plate is annular, the outer side of the floating ring plate is provided with a floating outer circumferential ring, and the inner side of the floating ring plate is provided with a floating inner circumferential ring; the floating outer circumferential ring is elastically connected with the top of the enclosing cylinder, and the floating inner circumferential ring is elastically connected with the tree base; the floating ring plate is provided with a water cavity which is concave downward, and a plurality of water permeable holes are arranged in the floating ring plate; a plurality of support strips are arranged at the bottom opening of the enclosing cavity, and the inner ends of the support strips are fixedly connected with the tree base.

This invention relates to an automated monitoring method for the dynamic changes of plant diversity in mountainous forest areas, comprising: constructing a three-dimensional model of the target mountainous area and deploying a sensor network; collecting microclimate data and soil characteristic data to determine a spatial distribution matrix; acquiring historical plant distribution data of the target mountainous area, mapping it to the spatial distribution matrix to construct a comprehensive distribution map of environmental factors, and determining a correlation model between plant distribution and environmental factors; acquiring real-time satellite remote sensing image data of the target mountainous area, using the correlation model to analyze the characteristics of changes in plant cover and plant diversity distribution, and locating several key plant distribution areas; identifying plant species and counting their numbers in the key plant distribution areas, updating the comprehensive distribution map of environmental factors, and completing the automated monitoring of the dynamic changes of plant diversity in the target mountainous area. This invention integrates multi-source environmental data to achieve automated analysis of plant diversity distribution.

The invention provides a self-circulation drinking water supply station for a mountain stream environment, which is characterized in that an integrated controller carries out cooperative management of power generation, energy storage and water treatment and supply, and comprises an integrated house, a horizontal shaft type wind driven generator, an inclined-jet type vertical hydraulic generator, a built-in toilet, an integrated working bar counter, an instant heating hanging type water dispenser and a glass door. Island operation of equipment is met through wind power and water power resources in a mountain stream, self-circulation utilization of mountain stream water is achieved by combining with built-in water treatment corollary equipment, and the problem of high-quality drinking water and beverage water resource supply of an original mountain forest is solved; and meanwhile, due to an integrated modular factory product structure, the application scene is greatly improved, the construction standard is reduced, and the system can adapt to most scenic spot scenes.

The invention relates to the technical field of polygonatum sibiricum planting, in particular to an under-forest polygonatum sibiricum ecological planting device and method in a karst stony mountainous area. According to the technical scheme, the sealwort planting device comprises a planting and transporting assembly, and a planting distance adjusting assembly used for adjusting the sealwort planting range is installed on one side of the planting and transporting assembly. According to the invention, the control unit controls the seedling storage pipe to release or close the polygonatum sibiricum seedlings in the four seedling transport pipes, and the planting spacing of the polygonatum sibiricum seedlings is adjusted according to the regional state of the karst rocky mountain forest, so that the planting spacing is well adjusted according to environmental factors; compared with traditional small and medium-sized planting equipment which needs to frequently change positions for planting polygonatum sibiricum, the equipment has higher efficiency and smaller size.

1. Name of the Design Product: Dual-purpose Brooch and Pendant (Mountain Forest Little Fox). 2. Use of the Design Product: Jewelry for wearing. 3. Design Key Points of the Design Product: Lies in the shape. 4. Picture or Photograph that Best Illustrates the Design Key Points: Perspective View 1.

The invention relates to the technical field of fire distance measurement, and provides a mountain forest fire distance accurate measurement method and system based on three-dimensional panorama. The method comprises the following steps: determining a preset mountain forest monitoring area, performing pitching image acquisition in the area, and identifying mountain forest topographic features; dual-spectrum low-power-consumption pan-tilt modules are obtained, monitoring range penetration fitting is executed in combination with topographic features, monitoring points and a monitoring area are generated, and each monitoring point is provided with one pan-tilt module; performing visible light and thermal imaging time-space synchronous acquisition on a monitoring area through a holder module of a monitoring point, calculating the distance from a fire point to the monitoring point, and generating a fire point distance measurement time sequence; future environment prediction information is collected, the fire point diffusion direction and speed are predicted in combination with a fire point distance measurement time sequence, a diffusion prediction result is generated, and the technical problems that an existing mountain forest fire monitoring system is poor in adaptability to complex terrains, low in distance measurement precision and insufficient in fire point diffusion prediction capacity are solved.

The present invention provides a method for remote sensing classification of forest types based on historical sample migration, belonging to the field of forest monitoring. The method first determines the monitoring area, historical years and target year, obtains the corresponding sample points, then obtains the long-term multi-source remote sensing data of the historical sample points and characterizes the forest growth and succession characteristics, then determines whether the forest type has changed, and migrates the sample data that has not changed to the target year; extracts classification features related to remote sensing classification using the multi-source remote sensing data of the target year; at the same time, divides the monitoring area into multiple floristic regions, and selects the optimal classification features for each floristic region from the relevant classification features; finally, classifies the forest types of each floristic region based on the optimal classification features and the migrated historical forestry field survey sample point data that have not changed. The present invention increases the number of sample points in the target year and improves the accuracy of remote sensing classification of mountain forest types under the condition of small samples.

The utility model is suitable for the technical field of gastrodia elata planting, and provides a water accumulation prevention device for gastrodia elata planting under a mountain forest. The drainage groove is formed in one side of the retaining wall and used for drainage of the gastrodia elata planting land under the mountain forest; the overflow pipe is fixedly mounted on the retaining wall, and the water outlet end of the overflow pipe extends into the drainage groove; the mesh enclosure is fixed at the water inlet end of the overflow pipe and is used for blocking impurities; the cover plate is arranged on the drainage groove; and the scraping plate is arranged in the drainage groove and is used for cleaning impurities. According to the water accumulation prevention device for planting gastrodia elata under the mountain forest, soil loss is reduced through the retaining wall, and the silt accumulation risk of the drainage groove is reduced from the source; the overflow pipe is matched with the drainage channel to form a drainage channel and matched with the mesh enclosure to intercept impurities to guarantee smooth drainage.

A weather forecast service integrated display board relates to the field of scenic spot weather forecast and is used for providing more visual weather forecast for mountain forest type scenic spots. The weather forecast service integrated display board comprises a pre-embedded base, a supporting column, a display screen and a broadcasting assembly. The embedded base is embedded underground; the supporting column is fixedly connected with the pre-buried base, part of the supporting column is buried underground, and the top of the supporting column forms an inclined plane; the display screen is arranged on the inclined face of the supporting column and attached to the inclined face of the top face of the supporting column, the edge of the display screen extends out of the supporting column, the display screen comprises a map part, a weather information part and a precautions part, the map part and the weather information part are distributed at intervals, and the map part and the precautions part are distributed at intervals. The announcement part is adjacent to the weather information part; the broadcasting assembly is arranged on the display screen, and the broadcasting assembly is arranged on the part, located outside the supporting column, of the display screen. The method is used for providing more visual weather forecast for tourists in the mountain forest type scenic spot.

The application discloses a mountainous forest tree root continuous drip irrigation system based on water storage irrigation, relates to the technical field of forest tree water storage irrigation, and comprises a tree hoop assembly arranged on the trunk of a forest tree, a water storage and flow guide assembly and a rainwater collecting and water guiding assembly. The tree hoop assembly is composed of a first hoop body and a second hoop body which are assembled by bolts. The water storage and flow guide assembly is fixed on the second hoop body. The tree hoop assembly, the water storage and flow guide assembly and the rainwater collecting and water guiding assembly are arranged to collect rainwater on site, store and purify the rainwater in the water storage and flow guide assembly, fix the whole system, stabilize the positions of the components, quantitatively store water, control the speed of water delivery, and deliver the water to the deep drip irrigation of the roots of the forest tree. The whole system integrates rainwater collection, water filtration, water storage and adjustable drip irrigation, does not need to lay a large-area field pipe network, is suitable for scattered forest tree planting in mountains, can be used for self-irrigation by rainwater in rainy days and external water supplement in dry seasons, and has the advantages of water saving, efficiency improvement and reduced cost of irrigation pipe protection in mountains.

The utility model relates to the technical field of unmanned aerial vehicle fire fighting, in particular to an unmanned aerial vehicle for mountain forest fire extinguishing. Comprising an unmanned aerial vehicle body, the bottom of the unmanned aerial vehicle body is provided with a detachably-connected suspension release device, the suspension release device comprises a bearing support seat detachably connected with the bottom of the unmanned aerial vehicle body, the bearing support seat is provided with a plurality of releasers, a control device and a water storage barrel, and the releasers are provided with detachably-connected suspension ropes. One end of each hanging rope is connected with the water storage barrel, and the control device controls the hanging ropes on the two adjacent releasers to be released to automatically disengage, so that the water storage barrel is turned over and toppled over when the hanging ropes are subjected to downward force. Therefore, the control device can accurately control the release of the hanging rope hung on the hook of the releaser, the accurate putting of a water source of the water storage barrel can be realized, and the fire extinguishing effect is improved; meanwhile, personnel are prevented from directly entering a dangerous area, and personnel safety is guaranteed.

The invention relates to the technical field of emergency rescue communication, in particular to a comprehensive positioning and emergency communication method for mountain jungle disaster rescue. Comprising the following steps: deploying an individual beacon and signal emission; deploying an unmanned aerial vehicle aerial platform and scanning response; sAR-based high-precision three-dimensional positioning of a complex environment is realized; constructing a self-adaptive 5G and Mesh ad hoc network based on perception; unmanned aerial vehicle precise hovering and communication guarantee based on positioning information; large-range communication coverage is carried out through multi-machine cooperation; through the above mode, the purposes of overcoming a complex terrain environment, realizing accurate positioning of a fire-fighting individual soldier, rapidly establishing a large-range and high-reliability emergency communication link, and guiding the unmanned aerial vehicle to carry out accurate communication guarantee are achieved.

The invention discloses a mountain area remote sensing image-based forest type identification method, and relates to the technical field of forest type identification. The method comprises the steps that a mountainous area remote sensing image to be recognized is cut to obtain a plurality of sub-image blocks, the space coordinate index of each sub-image block is recorded, and the cutting overlapping rate is larger than a preset value; respectively inputting each sub-image block into a trained mountain forest category identification model, and outputting a forest type classification result of each sub-image block; splicing the classification results based on the space coordinate index, and mapping the splicing results back to the mountain area remote sensing image space to be identified to obtain a forest type classification map with geographic reference; the model comprises a first classifier, and a multi-scale feature extraction unit, a feature fusion unit and a second classifier which are connected in sequence, and when the model is trained, the second classifier is assisted by the first classifier, so that loss values jointly participate in back propagation in a weighted sum mode. The method can improve the classification precision and classification efficiency of the forest types in the mountain area remote sensing image.