【摘要】：神經(jīng)網(wǎng)絡(luò)一直以來都是學(xué)術(shù)界研究的熱點(diǎn),而伴隨著圖形硬件的更新?lián)Q代,目前基于深度學(xué)習(xí)的神經(jīng)網(wǎng)絡(luò)再次在各個(gè)領(lǐng)域取得豐碩成果。然而這些人工神經(jīng)網(wǎng)絡(luò)處理信息時(shí)并沒有完整的考慮到生物神經(jīng)元的運(yùn)行機(jī)制,而作為神經(jīng)科學(xué)領(lǐng)域最新研究成果的Spiking神經(jīng)網(wǎng)絡(luò)具有高度的生物仿真性,其能夠很好的處理時(shí)空域維度上的特性信息,并將外界刺激以時(shí)間為特征進(jìn)行編碼,最后將編碼后的脈沖信息傳入神經(jīng)系統(tǒng)進(jìn)行處理,在生物身份認(rèn)證、語音識(shí)別等領(lǐng)域取得了很多的實(shí)際應(yīng)用成果。大腦生物神經(jīng)元能夠識(shí)別具有脈沖特性刺激信息的時(shí)序,然而這種識(shí)別機(jī)制的原理并沒有得到很好的揭示。研究基于Spiking神經(jīng)網(wǎng)絡(luò)的時(shí)序信息識(shí)別能更深入了解大腦對(duì)信息處理的原理,從而可將其識(shí)別原理應(yīng)用到識(shí)別處理外界復(fù)雜時(shí)空特性的信息,因此本文的研究非常具有科研前景。總的來說本文的研究?jī)?nèi)容如下:1.介紹了基于Spiking神經(jīng)網(wǎng)絡(luò)時(shí)序信息識(shí)別所涉及的相關(guān)基礎(chǔ)知識(shí),包括神經(jīng)元的相關(guān)生物特性和脈沖神經(jīng)網(wǎng)絡(luò)模型。同時(shí)從信息編碼、學(xué)習(xí)神經(jīng)元的訓(xùn)練和解碼時(shí)序神經(jīng)元模型的構(gòu)建三大模塊進(jìn)行相關(guān)理論的闡述。2.提出了一種新的監(jiān)督學(xué)習(xí)算法DL-PSD。針對(duì)脈沖神經(jīng)網(wǎng)絡(luò)中的時(shí)序信息識(shí)別,在經(jīng)典算法PSD基礎(chǔ)上,結(jié)合神經(jīng)元的時(shí)間延遲特性提出了DLPSD算法,提高了時(shí)序信息識(shí)別中項(xiàng)識(shí)別的效率。3.提出了一種改進(jìn)的脈沖神經(jīng)網(wǎng)絡(luò)序列解碼機(jī)制,構(gòu)建相應(yīng)的神經(jīng)元模型。傳統(tǒng)基于卷積的方式對(duì)時(shí)序信息的識(shí)別并沒有充分利用神經(jīng)元的生物特性,本文根據(jù)FSA識(shí)別手寫字的基本原理,結(jié)合了生物神經(jīng)元樹突存在雙穩(wěn)態(tài)平臺(tái)電壓的生物特性,構(gòu)建一種解碼特定時(shí)序的解碼單元模型4.最后將信息編碼、項(xiàng)識(shí)別以及時(shí)序信息解碼三大模塊構(gòu)成一個(gè)整體進(jìn)行時(shí)序信息的識(shí)別。相位編碼轉(zhuǎn)換圖像信息、DL-PSD訓(xùn)練學(xué)習(xí)神經(jīng)元完成項(xiàng)識(shí)別、新的解碼結(jié)構(gòu)模型識(shí)別特定的數(shù)字圖像序列。實(shí)驗(yàn)成功識(shí)別了特定的光學(xué)字符輸入序列,同時(shí)改變學(xué)習(xí)輸出神經(jīng)元與解碼模型感知單元的連接結(jié)構(gòu),可以識(shí)別出更多的特定數(shù)字序列,展示本文識(shí)別機(jī)制的可擴(kuò)展性和魯棒性,這對(duì)于構(gòu)建通用的識(shí)別結(jié)構(gòu)來編碼和處理人體生物特征信息提供了一條新的途徑,在圖像處理領(lǐng)域也有其應(yīng)用價(jià)值。
[Abstract]:Neural network has always been a hot topic in academic research. With the upgrading of graphics hardware, the neural network based on in-depth learning has once again achieved fruitful results in various fields. However, these artificial neural networks do not take into account the operation mechanism of biological neurons when processing information, and Spiking neural networks, as the latest research results in the field of neuroscience, have a high degree of biological simulation. It can deal with the characteristic information in the spatial dimension well, and encode the external stimulus with the characteristic of time. Finally, the encoded pulse information afferent neural system can be processed and authenticated in the biological identity. Speech recognition and other fields have achieved a lot of practical results. Brain biological neurons can recognize the timing of impulsive stimuli, but the principle of this recognition mechanism has not been well revealed. The research of time series information recognition based on Spiking neural network can better understand the principle of brain information processing, so it can be applied to the recognition and processing of complex temporal and spatial information. Therefore, the research of this paper is very promising. In general, the contents of this study are as follows: 1. This paper introduces the basic knowledge of temporal information recognition based on Spiking neural network, including the related biological characteristics of neurons and the model of impulsive neural network. At the same time, from the information coding, learning neuron training and decoding time series neuron model construction of three modules to explain the relevant theory. 2. A new supervised learning algorithm, DL-PSD., is proposed. In this paper, based on the classical algorithm PSD and the time delay characteristic of neurons, a DLPSD algorithm is proposed for the recognition of temporal information in impulsive neural networks, which improves the efficiency of item recognition in time series information recognition. An improved sequence decoding mechanism based on impulse neural network is proposed and the corresponding neuron model is constructed. The traditional method based on convolution does not make full use of the biological characteristics of neurons. According to the basic principle of FSA recognition and writing, this paper combines the biological characteristics of biological neuron dendrites with bistable plateau voltage. A decoding unit model for decoding specific timing is constructed. Finally, three modules, namely information coding, item recognition and timing information decoding, are integrated to recognize the timing information. DL-PSD trains learning neurons to complete item recognition, and a new decoding structure model is used to recognize specific digital image sequences. The experiment successfully identified a specific optical character input sequence, and changed the connection structure between the learning output neuron and the decoding model perception unit, so that more specific digital sequences could be identified. It shows the extensibility and robustness of the recognition mechanism in this paper, which provides a new way to code and process the biometric information of human body by constructing a universal recognition structure, and also has its application value in the field of image processing.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41;TP183

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 藺想紅;王向文;張寧;馬慧芳;;脈沖神經(jīng)網(wǎng)絡(luò)的監(jiān)督學(xué)習(xí)算法研究綜述[J];電子學(xué)報(bào);2015年03期

2 李婷;;國內(nèi)外神經(jīng)網(wǎng)絡(luò)的發(fā)展及概述[J];知識(shí)經(jīng)濟(jì);2013年18期

3 曹平;陳盼;章文彬;張潮;;基于脈沖神經(jīng)網(wǎng)絡(luò)的語音識(shí)別方法的初步探究[J];計(jì)算機(jī)工程與科學(xué);2008年04期

相關(guān)碩士學(xué)位論文 前1條

1 沈虹;Spiking神經(jīng)網(wǎng)絡(luò)及其應(yīng)用研究[D];南京工業(yè)大學(xué);2006年

，

本文編號(hào)：2216779