3S Technology In Forestl And Protection And Utilization Planning

Forestland is an important natural and strategic resource, which is the foundation for forestsurvival and development. Forestland plays a core role in maintaining land eco-security, and ithas a special status in response to global climate change. Forestland resources protection andutilization planning plays an important role in improving ecological environment, maintainingecological balance, and promoting social sustainable development of resources and environment.While the protection and utilization planning is an important means of overall planningforestland resources, building resource-saving and environment-friendly society, and enhancingthe ability of social and economic sustainable development.In this article, based on3S technology, using SPOT-5remote sensing data, and takingLongyuan Street of Longquan City in Zhejiang Province as the research object, after bandselection of multi-spectral data, image fusion between the multi-spectral bands and thepanchromatic band via a variety of methods, and evaluating the fused images were implementedto identify the optimal fusion results as the input of image classification. Then, the traditionalpixel-based image classification and the object-oriented classification were made respectively.The relevant data of the research area combined with the classified results were used to enableand facilitate the planning of forest protection and utilization. The conclusions of this researchwere summarized as follows:(1) According to the delineated results of the forestland boundaries in Zhejiang Province in2012, the current situation and problems of the forest protection and utilization in the study areawere analyzed. The statistics showed that forestland area was8278.6hectares. Forest area was7806.03hectares, accounting for94.29%of the forestland area. The forest coverage rate was76.90%.(2) Through qualitative analysis and the optimal index formula (OIF) calculation, theoriginal bands1、3and4were regarded as the best band combination. Based on SPOT-5multi-spectral bands and its panchromatic band, the Wavelet transform, HSV transform, PCAtransform and Gram-Schmidt transform were implemented to fulfill the operations of imagefusion, followed by a subjective qualitative evaluation coupled with quantitative objectiveevaluation of the fused results. The results showed that, Gram-Schmidt transform was the bestimage fusion algorithm.(3) In ENVI4.7environment, the maximum likelihood classification, the post supervisedclassification following the object-oriented classification and the decision trees classificationwere done respectively. The overall classification accuracy were estimated at87.71%for thedecision trees algorithm with a Kappa coefficient of0.8489,83.29%for the post supervisedclassification with a Kappa value of0.7948, and78.14%for the maximum likelihood classifier with a Kappa value of0.7318respectively. The results showed that compared with the traditionalpixel-based classification method, the object-oriented classification method has certainadvantages. Analysis of the vector data converted from the Object-oriented classificationsshowed that automatic computer based interpretation of the remotely sensed images greatlyimproves the work efficiency of delineating forest boundaries.(4) The goal of properly planning the forestland protection and utilization is to ensure thedynamic equilibrium of the forestlands, to promote a continuous growth of forest area, and tooptimize the structure of forest tree species in the study area. By dividing protection classes andquality levels of the woodlands, constructive recommendations have been proposed for a betterwoodland categorical management, systematic protection and management, quality gradingutilization management, origin structural optimization and protection and restoration for theforests. And the safeguard measures for forest protection and utilization in the study area werealso proposed.