閾值學習過程 的英文怎麼說

中文拼音 [yùzhíxuéxíguòchéng]
閾值學習過程 英文
threshold learning process
  • 閾 : 名詞1. [書面語] （門坎兒） threshold; doorsill2. （界限; 范圍） threshold
  • 學 : Ⅰ動詞1 （學習） study; learn 2 （模仿） imitate; mimic Ⅱ名詞1 （學問） learning; knowledge 2 （學...
  • 過 : 過Ⅰ動詞[口語] （超越） go beyond the limit; undue; excessiveⅡ名詞（姓氏） a surname
  • 程 : 名詞1 （規章; 法式） rule; regulation 2 （進度; 程序） order; procedure 3 （路途; 一段路） journe...
  • 過程 : process; procedure; transversion; plication; course
  1. We construct weak classifier by a haar feature ; then weak classifiers are combined to a strong classifier in a linear way. the final classifier is built in a cascade structure, which could reject most non - face samples in the early layer. also we use integral image to quickly calculate the feature and reduce the detection time

    本文以簡單的haar特徵結合閾值構造弱分類器，通過adaboost學習選擇和集成弱分類器，最後按照分層結構把集成的分類器組合在一起；同時，在檢測過程中採用積分圖的方法計算特徵，保證了檢測的速度。
  2. Before the bp neural net forecast fire size class, it needs a process of studying from sample data. the neural net adjusts the weight value and threshold value according to the sample so as to give the linking weight value and threshold to low the difference between output from itself and the expected value

    Bp網路在應用於預測預報之前，需要一個網路學習過程，網路根據輸入的訓練（學習）樣本進行自適應、自組織，確定各神經元的連接權w和閾值。
  3. During the course of develop fault diagnostic method, the influence to the training circle number with network structure 、 learning rate 、 initial weight value & door value etc are discussed. by comprehensive analyses and comparing, the comparatively rational value is adopted to be network ' s eigenvalue

    在制粉系統故障診斷樣本訓練過程中，本文作者探討了網路結構、學習率、初始權值閾值等因素對訓練速度的影響，為選取合理的網路參數提供了依據。
  4. Focusing on the difficulties in the knowledge sharing process, the author integrated the phenomenon of knowledge sharing and suggested using parameters such as knowledge gap, sharing threshold, sharing rate and self - studying to quantify the factors of knowledge sharing

    針對他們在知識共享過程中存在的困難，筆者結合常見的知識共享現象，提出用知識距離、共享閾值、共享比率以及自學習等參量將知識共享的影響因素加以量化。
  5. Later on, after elaborating the disadvantages of the old methods in detecting and recognizing moving objects, a series of corresponding approaches are proposed, such as grid scan, local tracking bug and dynamic window in object tracing to reduce the huge data needed to be processed, maximum and minimum for selecting a proper segmentation threshold and improved conversion from rgb model to hsv and so on to decrease the influence of inhomogeneous lighting and the color noise, a bilinear interpolation in each quadrant to eliminate the bad effect on the recognition precise because of the distortions of the camera. after that, much emphasis is given on application study in pattern recognition with a feed - forward neural network. both the basic bp algorithm and improved bp algorithm in the study process are described in detail, and the later is used to quicken convergence speed and improve validity of the network

    然後，分析和闡明了傳統的運動目標檢測方法的不足，並在此基礎上結合研究中的實際實驗環境，提出了一系列解決方法，包括針對降低龐大數據量而提出的網格掃描、局部「跟蟲」追蹤和動態窗口掃描等目標檢測方法，針對實驗環境中光照不均和顏色干擾提出基於人機交互的最大最小值閾值選取方法和引入改進的rgb模型到hsv模型的轉換方法，為消除圖像畸變對識別精度的惡劣影響而採用的通過控制點進行雙線性插值進行畸變校正的方法；緊接著，概述了神經網路的發展歷史和幾種常用神經網路模型的特點，重點研究了前饋型神經網路在模式識別中的應用問題，詳細闡述了基本的bp演算法和學習過程中bp演算法的改進，從而使網路收斂速度更快，解決問題更有效，並在此基礎上，設計了一個基於bp神經網路的運動目標識別系統，給出了實驗結果。
  6. ( 2 ) combining secondary genetic algorithm with back - propagation network, the thesis redacts genetic neural network procedure, which optimizes number of hidden node and weight value and threshold value simultaneously. the procedure overcomes blindness during search, avoids falling into localminimum and increases learning accuracy

    （ 2 ）編寫了遺傳神經網路ga - bp程序，採用二級遺傳演算法與bp演算法相結合，同時優化網路隱層節點數和權值、閾值，既克服了尋優過程的盲目性，又避免陷入局部極小，提高了網路的學習精度。
分享友人
分享到 Line

分享到臉書