新技术新工艺
新技術新工藝
신기술신공예
New Technology & New Process
2015年
10期
51-53,54
,共4页
烟丝图像%自适应消噪%模糊神经网络%在线检测
煙絲圖像%自適應消譟%模糊神經網絡%在線檢測
연사도상%자괄응소조%모호신경망락%재선검측
template cut tobacco image%adaptive noise cancellation%fuzzy neural network%on-line detection
针对目前烟丝在线检测设备进行图像检测识别时,测量结果容易受到噪声、光线、温度和湿度等的影响,而导致参数时变、时间滞后等问题,设计一种基于 MATLAB 软件的自适应消噪技术,利用自适应神经模糊推理系统(ANFIS)对烟丝图像噪声的非线性动态特性进行建模,并利用 ANFIS 逼近有色噪声,然后从测量烟丝图像中消除有色噪声,得到有用的烟丝图像。该技术在实际烟丝在线检测生产过程应用中表明,能够在复杂的环境下,自动地估计出噪声信号,方便、快捷地实现在线去噪,从而得到满意的烟丝图像。
針對目前煙絲在線檢測設備進行圖像檢測識彆時,測量結果容易受到譟聲、光線、溫度和濕度等的影響,而導緻參數時變、時間滯後等問題,設計一種基于 MATLAB 軟件的自適應消譟技術,利用自適應神經模糊推理繫統(ANFIS)對煙絲圖像譟聲的非線性動態特性進行建模,併利用 ANFIS 逼近有色譟聲,然後從測量煙絲圖像中消除有色譟聲,得到有用的煙絲圖像。該技術在實際煙絲在線檢測生產過程應用中錶明,能夠在複雜的環境下,自動地估計齣譟聲信號,方便、快捷地實現在線去譟,從而得到滿意的煙絲圖像。
침대목전연사재선검측설비진행도상검측식별시,측량결과용역수도조성、광선、온도화습도등적영향,이도치삼수시변、시간체후등문제,설계일충기우 MATLAB 연건적자괄응소조기술,이용자괄응신경모호추리계통(ANFIS)대연사도상조성적비선성동태특성진행건모,병이용 ANFIS 핍근유색조성,연후종측량연사도상중소제유색조성,득도유용적연사도상。해기술재실제연사재선검측생산과정응용중표명,능구재복잡적배경하,자동지고계출조성신호,방편、쾌첩지실현재선거조,종이득도만의적연사도상。
Currently when identify the cut tobacco image of on-line detection,the measurement results are easily influ-enced by noise,light,temperature and humidity,and which cause the parameter time-varying and time lag.Design the AN-FIS noise cancellation technology based on the MATLAB software.Modeling the nonlinear dynamic characteristic of noise by ANFIS and approaching the colored noise by ANFIS are presented.Then the colored noise is cancelled from measured signal and the useful cut tobacco image is gotten.This technology is applied in the actual cut tobacco image of on-line detec-tion production process,indicating that it can automatically estimate the noise signal,quickly and easily realize online noise cancellation,thus obtain the satisfied cut tobacco image.