Chung Cheng University Institutional Repository
Not a member yet
889 research outputs found
Sort by
考慮設置組合因素之隨機分析方法用於配電系統饋線最大允許太陽能發電滲透率評估;A Stochastic Analysis Method with Considering Installation Combination Factor to Evaluate the Maximum Allowable Penetration Level of Photovoltaic Generation on Distribution System Feeders
No full text[[abstract]]由於環保意識的抬頭，再生能源已經成為各國重要的發展項目之一。政府開始推動利用再生能源來發電，使用再生能源來發電可以保護地球、減少二氧化碳的排放。在眾多的再生能源之中，太陽能發電是世界上其中一個快速發展的項目之一，然而隨著日益增加的太陽能發電源併入電力系統，評估太陽能發電的潛在風險越顯重要，因此迫切地需要相關研究來探討饋線的太陽能承載容量，即饋線能接受且不會違反系統正常操作的太陽能最大併網裝置容量。太陽能發電系統併聯在不同位置或者不同容量會對饋線造成不一樣的衝擊，因此電力公司必須做一些積極的工作來探討之。 本論文提出設置組合因素之隨機分析方法來評估當太陽能系統被併聯在不同位置和容量以決定配電饋線的太陽能承載容量。考慮到太陽能系統併聯位置和容量的不可預測性，因此使用隨機分析法搭配排列組合方式來決定太陽能系統在饋線上所有不同的併聯位置組合數，再來慢慢遞增太陽能系統的容量來探討饋線上的狀況，並在太陽能模型中加裝智慧型變流器以改善饋線電壓變動問題和提升承載容量。配電饋線衝擊狀況會參考台灣電力公司的再生能源併聯要點並探討過電壓、電壓變動率與三相不平衡率三項指標，當分析出饋線上所有的電壓衝擊狀況，便可得知不會危害饋線穩定度和安全性時最線最大允許的太陽能承載容量。 本文使用OpenDSS和MATLAB進行共同模擬。OpenDSS主要用來建構饋線模型和執行電力潮流分析。而MATLAB用來計算出太陽能系統不同的併聯位置和容量，接著透過COM介面把位置與容量訊息傳送給OpenDSS，再透過OpenDSS所建構太陽能模型進行電力潮流求解，最後再由MATLAB紀錄所需資訊並產生分析結果。關鍵詞：太陽能承載容量、隨機分析法、智慧型變頻器控制、電壓變動
Owing to the increased environmental awareness grows in recent years. The government promotes the use of renewable energy to generate electricity. Using renewable energy to generate electricity can protect the earth and reduce carbon dioxide emissions. Among the numerous renewable energy sources, solar power is one of the rapid growth clean energy in the world. However, with the increasing number of solar power sources connected to the grid, it is increasingly important to evaluate the potential risks of solar power generation. So there is an urgent need for relevant research to do some work on hosting capacity of the feeder. The photovoltaic system will result to different impact when connecting at different sites or different capacity. Therefore, the power company must have to do something positive to discuss the issue. In this thesis, a stochastic analysis method with considering installation combination factor is use to evaluate the maximum allowable penetration level of photovoltaic generation on distribution system feeders. Consider the unpredictability of solar system connected sites and capacity for the reason that stochastic analysis with the combination factor is to use calculate total number of combinations of photovoltaic system, after that gradually increase the capacity of the photovoltaic system to explore the impact on the feeder. In addition, smart inverter strategy is adopted to improve the power quality and enhance the hosting capacity. The impact in distribution feeder will basis on the renewable energy connected rules of Taiwan power company for following criteria：overvoltage、voltage deviation and voltage unbalance. After getting all the voltage impact on the feeder, and we can know how amount capacity of photovoltaic system can be connected to the feeder and will not cause endanger to the grid. This thesis use OpenDSS and MATLAB co-simulation. The advantage of OpenDSS is to model the system and do the fast power flow calculation. It can not only run independently but also execute pass through COM interface and connect with other software. In this thesis, MATLAB is use to calculate different sites and different capacity of the photovoltaic system. After that pass the message to OpenDSS by COM interface to model the photovoltaic system and do the fast power flow calculation. Finally, MATLAB will collect needed data and depict.Keywords：Hosting capacity、Stochastic analysis、Smart inverter control、power quality criteri
液態金屬調控之可重置標籤電路設計;Design of Reconfigurable Transponder Circuit Based on Liquid Metal
No full text[[abstract]]本論文以液態金屬與軟性基板製作可重置電路，並討論通道製作與液態金屬的致動原理。PDMS軟性基板擁有形變的能力，且惰性與許多物質皆不會起反應，可耐熱耐酸鹼，是非常適合做為流道的材質。搭配液態金屬可形變與致動得特性，兩者相得益彰。此外由於液態金屬致動時能產生大幅度的物理形變，有利於突破半導體元件在高頻可調範圍下降導致重置性因此限縮的問題，使其重置性具有更大的調控空間。本論文以上述技術實現具振幅及相位調變之反射標籤，藉著通道的外形與致動方式的配合改變液態金屬形狀，使電路整體物理表面積改變，達到所需的效果，進而克服傳統以負載電路切換進行調變的限制。本論文另一主題為chipless RFID tag，以液態金屬可重置的特性，使被動電路具有調整狀態的優勢，能降低許多電路設計的成本與複雜度。本次設計利用多重諧振器實現chipless RFID tag，且具有3 bit的編碼能力，在量測時標籤與收發端距離可達60 cm。
This thesis focuses on the designs of microwave reconfigurable circuits using liquid metal and the flexible substrate. PDMS is adopted as flexible substrate due to numerous advantages, including high flexbility and low chemical reaction. On the other hand, the liquid metal can be physically deformed by applying proper actuation. The deformation can be dramatic, providing great reconfigurability comparing to other solid-state tuning devices.The first topic is to design a transponder. The liquid metal can physically change the shape and area of metal surface, resulting in amplitude and phase modulation. This proposed approach can increase RCS ratio coparing to the conventional switching-load technique.The second topic is chipless RFID tag. With liquid metal reconfigurability, the passive circuit has the advantage of adjusting the state, can reduce the cost and circuit design complexity. This design uses multiresonator to achieve chipless RFID tag, and has a 3 bit coding capability. In the measurement, the tag can be away from the transceiver up to 60 cm
基於快速探索隨機樹之自主移動機器人室內未知環境探索;Autonomous Mobile Robot Exploration in Unknown Indoor Environment Based on Rapidly-exploring Random Tree
No full text[[abstract]]近年來機器人迅速地融入人類的日常生活中，為了讓機器人能夠自主導航，必須提供準確的地圖，因此為了讓機器人能夠自動獲得地圖以及提升自主探索的效率，本篇論文提出了一套基於RRT與邊界的2D-SLAM探索系統，系統架構可分為三個部分，首先以雷射資訊建構出初始地圖，使用RRT與邊界探測器檢測初始地圖的邊界點。在第二部分我們將得到的所有邊界點經過過濾與聚類，減少邊界點的數量降低計算的需求。最後計算每個邊界點的效益，引導機器人前往未知區域，直到地圖建構完成。在實驗結果中，我們分別測試機器人在虛擬與實際環境中自主探索未知的室內場景，並評估了我們的系統在各種虛擬和實際的室內環境中的性能，實驗結果顯示，我們的系統能夠成功地找到未知區域，並且在合理的時間內完成自主探索。
In recent years, robots have quickly integrated into the daily life of people. In order to allow robots to navigate autonomously, accurate maps must be provided. Therefore, in order to enable robots to automatically obtain maps and improve the efficiency of autonomous exploration. This paper proposes a method based on RRT and frontier 2D-SLAM exploration system. The system architecture can be divided into three parts, firstly constructing the initial map with laser information. Using RRT and frontier detector to detect the boundary point of the initial map. In the second part, we will filter and cluster all the boundary points to reduce the number of boundary points and the computational. Finally, calculate the score of each boundary point and guide the robot to the unknown areas until the map is constructed. In the experimental results, we separately test the robots to explore unknown indoor environments in simulation and real environments. We evaluate the performance of our system in various simulation and real indoor environments. The experimental results show that our system can successfully find unknown areas and complete autonomous exploration in a reasonable time
國家級射箭教練領導行為及其影響因素之研究;Leadership Behaviors and Influential Factors among the National Archery Coaches
No full text[[abstract]]本研究目的在了解國家級射箭教練領導行為及其影響因素。採質性研究方式，選取四位訓練卓越的國家級射箭教練及其四位選手，教練本身具領導實務達10年以上，並且曾帶隊獲得國際賽前三名成績，目前仍然為射箭教練。本研究採多元教練領導行為理論為基礎，以半結構式訪談、參與觀察及文件分析等收集資料，所得資料以開放性編碼、主軸編碼及選擇性編碼，持續歸納、比較及分析，藉以正確且完整地詮釋資料的內涵與影響因素。結果發現，國家級射箭教練領導行為主要包括訓練與教學行為、關懷行為、專制行為、獎勵行為及民主行為等。教練的訓練與教學行為包括指導原則、訓練模式、個別指導；關懷行為包括心理輔導、課業輔導；專制行為包括服從態度、生活管理；獎勵行為包括精神鼓勵、實質的鼓勵；民主行為包括雙向互動、選手的意見。影響國家級射箭教練領導行為的因素包括訓練瓶頸、學校與家長的支持、訓練的資源及教練具備條件。教練的訓練與教學行為重視目標設定、鼓勵思考、多元化；訓練方法、心理訓練、比較法；技術動作的提升、個別化訓練、個別化指導。整體而言，教練重視運動的價值、衝突處理；升學輔導、補救教學；重視紀律與規範養成、彈性的空間；生活習慣養成、嚴格的紀律；重視言語讚賞、公開讚賞；金錢獎勵、禮物獎勵；偏差行為的處理、良好態度的鼓勵。重視建立溝通關係、信任關係、以及訓練感受的建議。另外，教練最需要具備的是對於運動訓練與指導的熱忱，吸收新的知識及自我要求等。教練是選手學習的榜樣，如果教練擁有好的專業知識及品性，都有可能間接或直接影響選手，讓選手變得更加出色。
The aim of the current study was to investigate the leadership behaviors and influential factors among the national archery coaches. Four professional archery coaches who were from national team, and four archery athletes from their team were participated in this study. All of the coaches have given archery training at least ten years and still given now, in addition, they have shepherded the national teams to top three in the international competition. A semi-structured interview that based on the qualitative method and multiple coaches' leadership behavior theory was used in this research, the collected interviewing data was transferred to the coding by open coding, axial coding, and selective coding. Furthermore, we induct, compare, and analyze for obtained the integrality of data and influential factors. The results demonstrated that leadership behaviors of national coaches including the training and teaching behavior, caring behavior, autocratic behavior, reward behavior, and democratic behavior etc.; the training and teaching behavior containing guiding principles, training model, and individual coaching; the caring behavior comprising psychological counseling and academic guidance; the autocratic behavior including obedience attitude as well as life management; the reward behavior consists of spiritual encouragement and material encouragement; whereas the democratic behavior including bilateral interaction as well as opinions from athletes. In addition, the influential factors of leadership behavior in national archery coaches including the bottleneck of training, support from school and paterfamilias, training resources, and qualified requirement of coaches. The training and teaching behavior emphasized the goal setting, encouragement for consideration, diversification; methods of training, psychological training, comparison; increasing movement skills, individually training, and individualy coaching. In overall, the coaches would pay more attention to the value of sport, conflict processing; careers counseling, remedial education; nurturance of discipline and norm, flexible space; nurturance of lifestyle, strict discipline; verbal reward, public reward; monetary reward, gift reward; deal of deviance, encouragement of good attitude. To rebuild the relationship of communication and trust, as well as the advices of experience on training.Finally, the most important factor is the passion of athletic training and coaching, obtain new knowledge, and self-critical requirement. Coaches showcased the model for athletes, if they have better professional knowledge and morality, if would directly or indirectly affect athletes, give them rise to excellent performance.The current study was to revealed the meaning of leadership behaviors from national coaches based on the qualitative methods, then provide a guideline for archery coaching in Taiwan
使用隱藏式馬可夫模型，建立智慧型手機APP使用者行為預測系統;Using hidden Markov chain to predict User's behavior for Apps
No full text[[abstract]]智慧型手機遊戲APP在Google的Google Play以及Apple的App Store兩平台之APP數量數量繁多，且遊戲類型APP競爭尤其激烈，因此APP的開發者莫不希望能夠儘量留住客戶，以創造最大的商業利益，因此若能預測可能流失用戶之情形，就能制定相關對策，降低顧客流失之速度。本研究以遊戲APP之使用者活動記錄為例，使用隱藏式馬可夫模型，建立依據時間序列所產生的使用者行為資料之使用者預測系統，藉由以該APP歷史資料所計算出之移轉機率矩陣與狀態值轉換機率，提供隱藏式馬可夫模型之訓練，並利用Viterbi演算法預測路徑之最終值為預測之狀態，建立使用者狀態之預測系統。實驗設計以用戶的持續活躍與可能流失之兩種情形為可能轉換之狀態，並依照遊戲APP使用者經常性會產生的活動為特徵值，進行模型的訓練與可能狀態之預測，提供APP的開發者可能流失用戶之名單，並採用十折交叉驗證與混亂矩陣之驗證方法，驗證系統之有效性，成功提供一個使用者可能行為狀態之預測系統。
Nowadays, you can find a lot of applications in Google’s Google Play and Apple’s App Store, and most of them are game applications. Therefore, all application developers are trying their best to keep users staying in application for developers’ interest. If we can predict the possible situation of losing users, we can develop countermeasures to slow down the pace.In this paper, we recorded users’ activities for example. Using Hidden Markov Model to build a prediction system by users’ behavior data generated from time series. By using the log of this application, to calculate the Transition probability Martix and Emission probability. Training Hidden Markov Model, and using the final route predicated by Viterbi algorithm as the predicated status to create a user prediction system.This experiment is base on two circumstance; users stay active and probably leave. According to game application users’ regular behaviors as the eigenvalues to train the modle and predicte the possible status, in order to provide a list of users that could possibly leave. Also, by using 10-Fold Cross Validation and Confusion Matrix authentication method to verify the system, can provide an effective prediction system to predict users’ behaveors
建構薄膜電晶體液晶面板顯示器之面板工程One Drop Filling製程良率預測模式;Construction of thin film transistor LCD panel display end One Drop Filling Process Forecasting Mode
No full text[[abstract]]過去台灣引以為傲TFT-LCD光電產業，面對全球化的競爭者與產業特性快速輪動的景氣影響，正一步步的面臨產業生存危機，而國內各家業者無不卯足全力開發新的應用技術或是新式製程技術來應付這詭譎多變的產業，但正國內外各家廠商拚新式技術突破時，是否該反饋自我品質良率的精進提升。生產良率一直都是生產事業的主要競爭力，它攸關生產者各種成本的稼動率，也是企業經營重要成敗關健因素之一，因此如何提升良率產業競爭性，這是每個生產者所該面對重要的課題。因此TFT-LCD面板產業如上述所言面臨嚴峻挑戰，除了自我產品規格技術提升外，更應該從提升品質良率基本功夫紮根做起，而ODF製程更是TFT-LCD產業最末端的玻璃關鍵製程，生產的良率攸關整體TFT-LCD生產成本達5成以上。如何透過提昇製程良率來增加產業競爭力，藉由優化生產流程或機台參數的改善以提昇生產良率增加生產效能，才能在這艱苛的市場狀況下生存。本研究是藉由工業4.0的資料探勘技術，對於生產中小尺寸TFT-LCD面板公司，現有的實際製程參數設定資料，針對每道製程，分別以不同的屬性選取進行研究變項分析進而找出影響良率關鍵要素進行探討。首先從研究對象工程資料庫，所儲存的資料，由專家依經驗法則挑選針對研究議題的一組屬性集合，接下來由weka.attributeSelection.GainRatioAttributeEval模組挑選影響研究議題屬性變項，最後匯整參考文獻中影響研究議題良率屬性因子，再以這三種維度縮減方式所建構的研究變項資料為基礎。在所有的研究變項經過資料前處理完畢之後，分別將三組研究變項，透過資料探勘四種分類技術，再來進行影響良率關鍵要素預測的分析及探討影響屬性隱藏知識。關鍵詞：TFT-LCD面板ODF製程、工業4.0、維度縮減、資料探勘。
In the past decades, Taiwan was proud of TFT-LCD optoelectronics industry. Nowadays, facing international competitors and industry characteristics influenced by the economy of rapid globalization wheeled, it encounter survival crisis. For the crucial moment, various domestic industries are taking efforts to develop new the application of new technology or process technology to deal with this strange and varied industry. However, while domestic and foreign manufacturers are trying to breakthroughs, should the self-quality feedback to be considered to enhance the yield of production? Production yield is major competitiveness to production business. It not only influences utilization rate of various cost producers, but also roles the key business success factor. Therefore, how to enhance the competitiveness of the industrial yield is the important issue that every producer needs to face. Like discussed above, TFT-LCD panel industry is facing critical challenges. In addition to the technical specifications of self-promotion, to improve the quality basic yield should start from the fundamental. The ODF process is the extreme end of the TFT-LCD industry Glass critical process, and the production yield of TFT-LCD production influence overall cost more than 50%. How to increase industrial competitiveness by improving the process yield, improve the production yield and increase production efficiency by optimizing the production process or machine parameters is the key to survive in this strict and difficult market.This study is designed based on the data mining technology of industry 4.0. Small and medium size TFT-LCD panel manufacturer as target, through the current process parameters settings for each process, select study variables by different respectively attributes, and find out the key elements that impact the yield.First of all, getting data from researched object, select one set of attributes for the study target by Expert rule of thumb. Next, select impacting research topic attribute variables by weka.attributeSelection.GainRatioAttributeEval module. Finally, compile yield influenced factors from references, and then construct these three dimension reduction methods based on the result.After all research variables are pre-processed, take three sets of study variables, analyze and explore the impact of property hidden knowledge through data mining techniques four categories.Keywords: TFT-LCD panel ODF process, Industry 4.0, dimension reduction, data mining
基於PHF與MHOG混合特徵於行人檢測;Pedestrian Detection Based on PHF and MHOG Mixed Feature.
No full text[[abstract]]本研究提出一改進方法來解決單張影像行人檢測中，人潮擁擠或行人部分被遮蔽的情況。首先我們選用了兩種特徵萃取方法分別地使用於行人不同的部位，一是基於Gabor濾波前處理的MHOG特徵，此特徵萃取方法是延伸行人檢測的經典方法HOG，可有效增加此特徵對於行人上半身的描述；另一個是基於把原有的Haar-like特徵修改成適合行人部件使用的PHF特徵，因行走中的手臂與腿時常處於歪斜的狀態。兩種特徵萃取方法皆有使用積分圖加速法有效地加速運算。 得到行人特徵後，藉由兩層的SVM分類器來處理行人部分被遮蔽的情況，第一層SVM用來判斷各個部件是否被遮蔽，接著使用未遮蔽的部件所得到的機率分數作為對應的第二層SVM分類器的特徵值，即可判斷該檢測窗口是否有行人。最後使用ROC曲線及混淆矩陣兩個評估方式進行實驗，結果證明本研究提出之方法能有效處理人潮擁擠或行人部分被遮蔽的情況，並可以在不同的場景中成功地實現行人檢測。
Pedestrian detection is a considerable practical interest. The study proposes an improved methodology to solve crowded scenes or partial occlusion problem in pedestrian detection. First, the study used two feature extraction methods in different parts of pedestrians. One feature descriptors is MHOG based on Gabor filter and the other is PHF. They modified from HOG and Haar-like features descriptors and are adaptive for the shape of human upper body and human limbs, respectively. Both feature extraction methods use the integral image to effectively speed up the step. Second, the two-layer SVM classifier is used to deal with the partial occlusion problem. The first layer SVM can be used to determine that which human part is occluded, and then let the probability scores that obtained by the unoccluded parts as the feature value of the second SVM classifier. Through this processing, we can determine whether the detection window contains pedestrians. The experiment was tested by the ROC curve and the confusion matrix and the experiment result demonstrate that the method for pedestrian detection can effectively solve crowded crowd or partial occlusion problem, and it can be achieved in different outdoor environments
視覺慣性測量：用以強化即時定位與地圖構建;Toward Robust SLAM via Visual Inertial Measurement
No full text[[abstract]]為了實現機器人自主巡航，即時定位與地圖構建（SLAM 或 Simultaneous localization and mapping）是不可或缺的關鍵技術，過去二十年來的技術進展，二維 SLAM 己趨完備，如今為了讓機器人有更完整的感知，技術逐漸朝向三維稠密 SLAM 方向發展，期望機器人以此資訊與環境有更多元的互動。然而現有多數視覺三維 SLAM 系統並不夠強健，影像模糊、光度變化、低紋理場景都有可能使匹配失敗，並且傳統方法為了處理各種問題，使系統流程變得龐大複雜，而深度學習的發展帶來改變的契機。本研究以深度網路模型描述複雜的相機運動，有別於以往監督式學習VO研究，需要不易大量取得的相機移動軌跡，本研究提出了以影像輸入與 IMU 輸出作為 End-to-end 監督式學習訓練基礎之系統，使資料收集低成本化，光流架構也使系統不受訓練資料外觀影響，在與訓練資料完全不一樣的場景也能有正確預測。實驗結果顯示，本研究提出之架構比起同類研究有更快的訓練收斂速度，並且模型參數也大量減少，在比KITTI更具挑戰性的EuRoC資料集上也能正確預測，並且對於景像模糊、光度變化與低紋理場景都有一定的強健性，未來的研究能基於此成果，將預測與VIO系統整合，建構更強健的視覺SLAM系統。
To achieve autonomous robot navigation, SLAM (simultaneous localization and mapping) is a crucial component. For more than two decades of research, 2D-SLAM has already been deployed to the industry. Nowadays, we want the robot to have even more complete perception abilities, therefore more and more researchers put their effort on 3D dense SLAM.However, most of the existing visual 3D SLAM systems are not robust enough. Image blur, variation of illumination, and low-texture scenes may lead to registration failures. In order to deal with these problems, the work flow of traditional approaches become bulky and complex. On the other hand, the advancement of deep learning brings new opportunities.We use a deep network model to predict complex camera motion, which is different from previous supervised learning VO researches, and requires no camera trajectories that are difficult to obtain.Using image input and IMU output as end-to-end training pair makes data collection cost-effective. The optical flow structure also makes the system not depend on the appearance of training sets. The experimental results show that the proposed architecture has a faster training convergence than the similar research, and the model parameters are also greatly reduced. It can also correctly predict the EuRoC dataset that is more challenging than KITTI dataset. Our method could remain certain robustness under image blur, illumination changes and low-texture scenes. Based on this result, future research can integrate prediction with the VIO system to construct a more robust visual SLAM system
使用卷積碼並建構於可程式邏輯板之人體通道傳輸收發器之設計與實現;Design and Implementation of a Human Body Channel Communication Transceiver on FPGA Using Convolutional Codes.
No full text[[abstract]]現今，隨著技術的進步，穿戴式個人娛樂裝置與個人醫療照護裝置愈來愈普及。對於醫療照護裝置，傳統上，病人必須到醫院或者是在家接著線材監測生理狀況，對於常時間觀察上十分不方便，因此有許多無線的隨身監控的醫療裝置被應用，一般來說，無線的穿戴式裝置以藍芽、wifi來傳遞訊號，但是無線信號接收器有著高功耗、易受到干擾的問題。因此，由人體來當作信號傳輸媒介的技術被提出，因為信號在人體通道內具有較低的信號衰減且不容易受到周遭環境干擾的特性，人體通道傳輸能克服大部分使用無線射頻技術遇到的困難，另外，因為信號在離開人體耗能量會急遽衰減，因此，相對於使用無線射頻技術，人體通道傳輸較不會有個人隱私會被竊聽盜取的問題。本論文提出了一個寬帶式訊號使用卷積加密方法的人體通道傳輸器，並將其實作在可程式化邏輯版(FPGA)上，目的是為了快速驗證所提出架構的可行性。在傳收端，資料會由卷積加密器加密並且經過位元填充器和NRZI編碼後，藉由SMA線傳輸到人體上。在接收器端，一個過取樣的相位偵測器用來回復資料與時脈，由於透過卷積碼增加了資料的可靠性，在140公分的傳遞距離下使用3.125 Mbps( 6.25Mcps )的資料傳輸速度收送資料，bit error rate (BER)表現可以達到傳輸接收108個位元資料沒有任何錯誤，因此，所提出的人體通道接收器適用於醫療穿戴性裝置與娛樂性裝置上。
Nowadays, portable entertainment and healthcare devices has developed rapidly because of the advance semiconductor fabrication process. For the chronic patients, the wireless healthcare devices brings a lot of convenience. In general, the wireless radio frequency (RF) communication has drawbacks of high power consumption and sensitive to interferences. Body channel communication (BCC) uses the human body as a transmission medium which is more stable and has less power consumption. However, the body antenna effects causes interferences in human body.In this thesis, a wideband signal BCC transceiver using the convolutional encoding method is implemented on a field programmable gate array (FPGA) in order to verify the feasibility of the proposed design. In the transmitter side, the data will be encoded by convolutional encoding method and then modulated by a bit-stuffer and a NRZI modulator. The modulated data are transmitted directly to a human body through a SMA cable and an electrode. In the receiver side, the oversampling phase detection CDR will recover the clock and data. By means of using convolutional encoding method to increase the data reliability, there are no error bits occurred over 108 bits received at data rate 3.125 Mbps( 6.25Mcps ) in transmission distance under 140 cm. According to the experimental results, the proposed BCC transceiver is suitable for healthcare and entertainment devices
針對短波紅外線液晶可調式濾波器的光學瑕疵設計光譜影像校正方法;Design Spectral Image Correction Methods for Optical Defects in Shortwave Infrared Liquid Crystal Tunable Filters
No full text[[abstract]]不管在軍事上或是搜救上的目標物偵測應用，短波紅外線高光譜影像都能提供很詳細的光譜資訊。大部分的高光譜取像是透過推掃式方式對靜態場景進行掃描，而本研究則是使用短波紅外線相機搭配液晶可調式光學濾波器擷取動態場景的高光譜影像。由於蒐集到的影像中有干涉條紋，造成空間與頻譜上的失真而無法準確偵測目標物，本研究目的是要開發校正演算法以消除影像上的干涉條紋，並提高目標偵測率。經過一系列的實驗探討干涉條紋的形成原因，發現從硬體端排除此干擾成本極高，因此改採用演算法進行排除。我們蒐集多組參考影像，提出三種校正方法：(1)以模板為基礎的消去法，直接利用補正值校正。(2)兩點校正法，利用兩個基礎點計算出線性特徵曲線進行校正。(3)多項式擬合校正法，利用多個基礎點以多項式擬合計算出非線性的特徵曲線進行校正。我們發現最後一種方法效果最好。三種方法的技術細節和校正後的結果記載在本論文中。在物質辨識上，我們對於部分對焦影像與全對焦影像分別進行辨識，並使用了四種頻譜像素比對方法：(1) Spectral Angle Mapper；(2) Orthogonal projection divergence；(3) Orthogonal Subspace Projection；(4) Mahalanobis distance。最後利用不同方法辨識的結果比較及分析兩種影像。
Short-wave infrared (SWIR) hyperspectral image (HSI) provides detailed spectral information regardless of target detection applications in military or search and rescue. While most HSI tasks image a static scene by a push-broom scanner, here a liquid crystal tunable filter (LCTF) is mounted in front of a broad-band SWIR camera to take snapshots of dynamic scenes. Due to the optical defects in LCTF, the collected images have pattern of interference fringes (PIF), which causes spatial and spectral distortion and interferes the target recognition tasks. Therefore, the purpose of this study is to develop a correction algorithm to eliminate PIF in the spectral images, and improve the recognition accuracy.We experimentally investigated the cause of PIF, and found the cost of hardware solution is very high. Therefore, we turned to develop algorithmic solutions. By collecting the reference images with PIF only, we examined three correction methods: (1) Template-based elimination, which directly subtracts the reference from the collected image; (2) Two-point correction, which uses two referenced intensities to derive a linear curve to predict the original pixel intensities; (3) Polynomial fitting, which uses multiple reference intensities to derive a nonlinear curve to predict the original pixel intensities. We found the third method is the most effective. Their technical details and corresponding experimental results are documented in the thesis