脉冲耦合神经网络模型参数自动标定算法研究-电子科学与技术、电路与系统专业毕业论文.docxVIP

兰炳A-学硕±学位论文 兰炳A-学硕±学位论文 Abs仃act Artificial neural networks plays the key role in the fields of artificial intelligence and image processing because of its imitation of the wisdom and power of biological system．As new kind of neuraI network,Pulse-Coupled Neural Network (PCNN)has been put more emphasis on the research in many fields since it was proposed by Johnson in 1994．PCNN model was created by the research and understanding in the visual codex of small mammal such cat and produces synchronous bursts of pulses from with similar activi吼effectively grouping them by phase and pulse frequency,which makes it significantly difierent from the conventional artifieial neurons．Just because the model derives directly from the studies ofthe mammal’s visual properties．PCNN has been used successfully in image processing fields． PCNN model is algorithm wi血multiple parameters．and finding the proper value of these parameters is task．The drawback of the PCNN hinders its further development and application．So many simplified PCNN modeI is adopted in many papers．The remaining parameters are，nevertheless，manually adiusted according to the concrete application．The high accuracy is unachievable with the simplified modeI and the quality oflatter study is affected． Up to now．no Papers which investigate the auto—determination of the parameter values ofPCNN modeIin full．range and systematically exist．Based the mentioned above，two algorithms which performs the auto-determination presented in this research．As part ofthis paper,genetic algorithm which is general purpose stochastic opfim妇ion for search problem is utilized to determine these parameter values automatically．and automatic PCNN system is established．Through the segmentation．the performance ofthe automatic PCNN svstem is verified．As a／lother part of this paper,gradient descent algorithm is adopted to search parameters which reduce the between the desired output and the actLlal output盯aduaHy according to the LMS principle．then self-a