Opencv intro

Preview:

Citation preview

Intro to 影像處理 / 辨識 &

OpenCV kazami前影像處理 / 辨識工程師

我是誰?影像處理 / 辨識工程師大學:化工碩士:理論物理(廣義相對論 \ 重力波 \ 光應用物理 )

所以 ...

影像處理 / 辨識工作到底在幹嘛?

xxx Bank

一開始 , 大家認為影像處理是這樣

https://www.youtube.com/watch?v=csvt6JBAwBk

我以為影像處理是這樣

結果我在做這樣 ... Orz…

影像處理是什麼? 能賺錢嗎?Library工廠品保( PCB, ITO 玻璃 , 面板 , 造紙廠 , 花生 )特殊影像識別系統 ( 保全系統 , 指紋辨識系統 , 自動駕駛 )UI 介面 ( 辨識手勢 ...)

Computer Vision

CameraLightImage ProcessingRecognition

Camera- 劇情總是從出乎意料的地方開始Exposure Time: cmos sensor 接受光的時間ISO : 感光度Aperture : 光圈大小 (影響進光量)Frame Rate : 每秒拍攝的頻率

Camera

Light

Intensity :光的強度 (若光不夠可能會曝光不足)Frequency :光的頻段 (紫外光 可見光 紅外線 X光 微波)Color temperature :色溫 ( RGB 比例的偏重)

CCD 感光

Gray = 0.299 * Red + 0.587 * Green + 0.114 * Blue

以上絕對不是要拖台錢為什麼我用相同的演算法總是辨識錯誤?

為什麼昨天辨識成功今天就辨識不出來?

為什麼找到一堆不是我要的資訊?

Digital Image Processing

Pixel: the minimal element for an image

How to describe the information on a pixel?Color space. RGB, HIS, YCbCr, etc.

Digital Image Processing

Processing for what?EnhancementFeature extractionRecoveryRetouchingCompression

Related Libraries

HalconOpenCVSimpleCVEuresys

OpenCVOpenCV (Open Source Computer Vision Library) is an open source library.

computer visionmachine learning

The library has more than 2500 optimized algorithms.

OpenCVInterfaces:

C++CPythonJavaMATLAB

OS:WindowsLinuxAndroidMac OS

OpenCVModules:

corebasic data structuresfile I/O

imgprocimage filteringgeometrical image transformationscolor space conversion

Videomotion estimationobject tracking

OpenCVModules:

calib3dmultiple-view geometry algorithmsstereo correspondence algorithms

features2dfeature detectorsfeature descriptorsfeature matchers

highguivideo capturingimage and video codecssimple UI

GPU (Cuda)

Image Processing

Blurring convolutionkernel(Average)

FaceDetecttrained with particular object (positive example)arbitrary images of the same size (negative example)boosted: create Strong classifier which is a weighted sum of these weak classifierscascade: resultant classifier consists of several simpler classifiers (stages)

Features20*20 demo picturesFor each feature, it finds the best threshold which will classify the faces to positive and negative

C

CvHaarClassifierCascadeCvHaarStageClassifier

CvHaarClassifierCvHaarFeatur

e

CascadeCascade:

Stage1:Classifier11

Feature11Classifier12

Feature12…..

Stage2:Classifier21

Feature21Classifier22

Feature22…..

Machine Learning

Why miscalculations ?training stagehatglassessmileposture

檢測情境的差異Facebook

品質高自動對焦後製

工業手動對焦手動調光速度快精度高

Q&A

Recommended