区间二型模糊神经网络的遥感影像分类  

Interval Type-2 Fuzzy Neural Network for Remote Sensing Image Classification

作　　者：桂琪皓 王春艳 GUI Qihao;WANG Chunyan(School of Software,Liaoning Technical University,Huludao,Liaoning 125105,China)

机构地区：[1]辽宁工程技大学软件学院,辽宁葫芦岛125105

出　　处：《遥感信息》2024年第6期130-138,共9页Remote Sensing Information

基　　金：国家自然科学基金青年科学基金(41801368);辽宁省教育厅基础研究青年项目(LJKOZ2021154)。

摘　　要：针对遥感影像分类中的不确定性挑战,提出了一种创新的区间二型模糊神经网络。算法整合了区间二型模糊逻辑系统(interval type-2 fuzzy logic system,IT2FLS)和神经网络的优势,以增强模型处理不确定性的能力及其自适应特征学习性能。通过双重模糊器配置,包括模糊隶属函数参数的区间二型模糊器和基于嵌套的区间二型模糊器,精确捕捉遥感影像中的不确定性因素。新加入的模糊规则库和推理机减少了对模糊器中先验知识的依赖,同时增强了模型的鲁棒性。在DLRSD数据集中针对网球场场景的实验表明,与现有的模糊神经网络方法相比,该算法在分类准确率上提高了14.77%。在WHDLD数据集上的测试也显示出5.11%的性能提升,证明了该模型的有效性和优越性。In response to the uncertainty challenges in remote sensing image classification,this paper proposes an innovative interval type-2 fuzzy neural network.The algorithm integrates the advantages of interval type-2 fuzzy logic systems and neural networks to enhance the model’s capability to handle uncertainties and its adaptive feature learning performance.By employing a dual fuzzifier setup,including an interval type-2 fuzzifier for fuzzy membership function parameters and a nested interval type-2 fuzzifier,the system accurately captures uncertainty factors in remote sensing images.The newly added fuzzy rule base and inference engine reduce dependence on prior knowledge within the fuzzifiers while enhancing the model’s robustness.Experiments on the DLRSD dataset,targeting tennis court scenarios,show that the algorithm improves classification accuracy by 14.77% compared to that of the existing fuzzy neural network methods.Performance tests on the WHDLD datas et al so demonstrate a 5.11%improvement,verifying the effectiveness and superiority of the proposed model.

关 键 词：模糊神经网络 模糊逻辑系统 不确定性建模 双重模糊器 遥感影像分类 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置] O159[自动化与计算机技术—控制科学与工程]