private void Start()
{
webCamTextureToMatHelper = gameObject.GetComponent <WebCamTextureToMatHelper>();
webCamTextureToMatHelper.Initialize();
grayMat = new Mat();
makerGrayMat = new Mat(originMakerTexture.height, originMakerTexture.width, CvType.CV_8UC1);
makerTexture = new Texture2D(originMakerTexture.width, originMakerTexture.height);
Graphics.CopyTexture(originMakerTexture, makerTexture);
detector = ORB.create();
extractor = ORB.create();
// Get Key Points of Maker
makerMat = new Mat(originMakerTexture.height, originMakerTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(makerTexture, makerMat, false);
makerKeyPoints = new MatOfKeyPoint();
makerDescriptors = new Mat();
Imgproc.cvtColor(makerMat, makerGrayMat, Imgproc.COLOR_BGR2GRAY);
detector.detect(makerGrayMat, makerKeyPoints);
extractor.compute(makerGrayMat, makerKeyPoints, makerDescriptors);
matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
}
/// <summary>
/// Initializes a new instance of the <see cref="PatternDetector"/> class.
/// </summary>
/// <param name="detector">Detector.</param>
/// <param name="extractor">Extractor.</param>
/// <param name="matcher">Matcher.</param>
/// <param name="ratioTest">If set to <c>true</c> ratio test.</param>
public PatternDetector(ORB detector, ORB extractor, DescriptorMatcher matcher, bool ratioTest = false)
{
if (detector == null)
{
detector = ORB.create();
detector.setMaxFeatures(1000);
}
if (extractor == null)
{
extractor = ORB.create();
extractor.setMaxFeatures(1000);
}
if (matcher == null)
{
matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMING);
}
m_detector = detector;
m_extractor = extractor;
m_matcher = matcher;
enableRatioTest = ratioTest;
enableHomographyRefinement = true;
homographyReprojectionThreshold = 3;
m_queryKeypoints = new MatOfKeyPoint();
m_queryDescriptors = new Mat();
m_matches = new MatOfDMatch();
m_knnMatches = new List <MatOfDMatch>();
m_grayImg = new Mat();
m_warpedImg = new Mat();
m_roughHomography = new Mat();
m_refinedHomography = new Mat();
}
// Use this for initialization
void Start()
{
Texture2D imgTexture = Resources.Load("lena") as Texture2D;
Mat img1Mat = new Mat(imgTexture.height, imgTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture, img1Mat);
Debug.Log("img1Mat dst ToString " + img1Mat.ToString());
Mat img2Mat = new Mat(imgTexture.height, imgTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture, img2Mat);
Debug.Log("img2Mat dst ToString " + img2Mat.ToString());
float angle = UnityEngine.Random.Range(0, 360), scale = 1.0f;
Point center = new Point(img2Mat.cols() * 0.5f, img2Mat.rows() * 0.5f);
Mat affine_matrix = Imgproc.getRotationMatrix2D(center, angle, scale);
Imgproc.warpAffine(img1Mat, img2Mat, affine_matrix, img2Mat.size());
FeatureDetector detector = FeatureDetector.create(FeatureDetector.ORB);
DescriptorExtractor extractor = DescriptorExtractor.create(DescriptorExtractor.ORB);
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
Mat descriptors1 = new Mat();
detector.detect(img1Mat, keypoints1);
extractor.compute(img1Mat, keypoints1, descriptors1);
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
Mat descriptors2 = new Mat();
detector.detect(img2Mat, keypoints2);
extractor.compute(img2Mat, keypoints2, descriptors2);
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
MatOfDMatch matches = new MatOfDMatch();
matcher.match(descriptors1, descriptors2, matches);
Mat resultImg = new Mat();
Features2d.drawMatches(img1Mat, keypoints1, img2Mat, keypoints2, matches, resultImg);
Texture2D texture = new Texture2D(resultImg.cols(), resultImg.rows(), TextureFormat.RGBA32, false);
Utils.matToTexture2D(resultImg, texture);
gameObject.GetComponent <Renderer> ().material.mainTexture = texture;
}
public bool descriptorsORB_Old(Mat RGB, Mat cameraFeed, string targetName)//找出特徵的顏色方法三(可運行但效率不佳放棄)
{
if (RGB == null)
{
Debug.Log("RGB Mat is Null");
return(false);
}
//將傳入的RGB存入Src
Mat SrcMat = new Mat();
RGB.copyTo(SrcMat);
//比對樣本
Texture2D imgTexture = Resources.Load(targetName) as Texture2D;
// Texture2D imgTexture2 = Resources.Load("lenaK") as Texture2D;
//Texture2D轉Mat
Mat img1Mat = new Mat(imgTexture.height, imgTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture, img1Mat);
//創建 ORB的特徵點裝置
FeatureDetector detector = FeatureDetector.create(FeatureDetector.ORB);
DescriptorExtractor extractor = DescriptorExtractor.create(DescriptorExtractor.ORB);
//產生存放特徵點Mat
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
Mat descriptors1 = new Mat();
MatOfKeyPoint keypointsSrc = new MatOfKeyPoint();
Mat descriptorsSrc = new Mat();
//找特徵點圖1
detector.detect(img1Mat, keypoints1);
extractor.compute(img1Mat, keypoints1, descriptors1);
//找特徵點圖Src
detector.detect(SrcMat, keypointsSrc);
extractor.compute(SrcMat, keypointsSrc, descriptorsSrc);
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
MatOfDMatch matches = new MatOfDMatch();
matcher.match(descriptors1, descriptorsSrc, matches);
DMatch[] arrayDmatch = matches.toArray();
for (int i = arrayDmatch.Length - 1; i >= 0; i--)
{
// Debug.Log("match " + i + ": " + arrayDmatch[i].distance);
}
//做篩選
double max_dist = 0;
double min_dist = 100;
//-- Quick calculation of max and min distances between keypoints
double dist = new double();
for (int i = 0; i < matches.rows(); i++)
{
dist = arrayDmatch[i].distance;
if (dist < min_dist)
{
min_dist = dist;
}
if (dist > max_dist)
{
max_dist = dist;
}
}
Debug.Log("Max dist :" + max_dist);
Debug.Log("Min dist :" + min_dist);
//只畫好的點
List <DMatch> matchesGoodList = new List <DMatch>();
for (int i = 0; i < matches.rows(); i++)
{
//if (arrayDmatch[i].distance < RateDist.value * min_dist)
//{
// //Debug.Log("match " + i + ": " + arrayDmatch[i].distance);
// matchesGoodList.Add(arrayDmatch[i]);
//}
}
MatOfDMatch matchesGood = new MatOfDMatch();
matchesGood.fromList(matchesGoodList);
//Draw Keypoints
Features2d.drawKeypoints(SrcMat, keypointsSrc, SrcMat);
//做輸出的轉換予宣告
Mat resultImg = new Mat();
// Features2d.drawMatches(img1Mat, keypoints1, SrcMat, keypointsSrc, matchesGood, resultImg);
List <Point> P1 = new List <Point>();
// List<Point> P2 = new List<Point>();
List <Point> pSrc = new List <Point>();
Debug.Log("MatchCount" + matchesGoodList.Count);
for (int i = 0; i < matchesGoodList.Count; i++)
{
P1.Add(new Point(keypoints1.toArray()[matchesGoodList[i].queryIdx].pt.x, keypoints1.toArray()[matchesGoodList[i].queryIdx].pt.y));
pSrc.Add(new Point(keypointsSrc.toArray()[matchesGoodList[i].trainIdx].pt.x, keypointsSrc.toArray()[matchesGoodList[i].trainIdx].pt.y));
//Debug.Log("ID = " + matchesGoodList[i].queryIdx );
//Debug.Log("x,y =" + (int)keypoints1.toArray()[matchesGoodList[i].queryIdx].pt.x + "," + (int)keypoints1.toArray()[matchesGoodList[i].queryIdx].pt.y);
//Debug.Log("x,y =" + (int)keypoints2.toArray()[matchesGoodList[i].trainIdx].pt.x + "," + (int)keypoints2.toArray()[matchesGoodList[i].trainIdx].pt.y);
}
MatOfPoint2f p2fTarget = new MatOfPoint2f(P1.ToArray());
MatOfPoint2f p2fSrc = new MatOfPoint2f(pSrc.ToArray());
Mat matrixH = Calib3d.findHomography(p2fTarget, p2fSrc, Calib3d.RANSAC, 3);
List <Point> srcPointCorners = new List <Point>();
srcPointCorners.Add(new Point(0, 0));
srcPointCorners.Add(new Point(img1Mat.width(), 0));
srcPointCorners.Add(new Point(img1Mat.width(), img1Mat.height()));
srcPointCorners.Add(new Point(0, img1Mat.height()));
Mat originalRect = Converters.vector_Point2f_to_Mat(srcPointCorners);
List <Point> srcPointCornersEnd = new List <Point>();
srcPointCornersEnd.Add(new Point(0, img1Mat.height()));
srcPointCornersEnd.Add(new Point(0, 0));
srcPointCornersEnd.Add(new Point(img1Mat.width(), 0));
srcPointCornersEnd.Add(new Point(img1Mat.width(), img1Mat.height()));
Mat changeRect = Converters.vector_Point2f_to_Mat(srcPointCornersEnd);
Core.perspectiveTransform(originalRect, changeRect, matrixH);
List <Point> srcPointCornersSave = new List <Point>();
Converters.Mat_to_vector_Point(changeRect, srcPointCornersSave);
if ((srcPointCornersSave[2].x - srcPointCornersSave[0].x) < 5 || (srcPointCornersSave[2].y - srcPointCornersSave[0].y) < 5)
{
Debug.Log("Match Out Put image is to small");
SrcMat.copyTo(cameraFeed);
SrcMat.release();
Imgproc.putText(cameraFeed, "X-S", new Point(10, 50), 0, 1, new Scalar(255, 255, 255), 2);
return(false);
}
// Features2d.drawMatches(img1Mat, keypoints1, SrcMat, keypointsSrc, matchesGood, resultImg);
Imgproc.line(SrcMat, srcPointCornersSave[0], srcPointCornersSave[1], new Scalar(255, 0, 0), 3);
Imgproc.line(SrcMat, srcPointCornersSave[1], srcPointCornersSave[2], new Scalar(255, 0, 0), 3);
Imgproc.line(SrcMat, srcPointCornersSave[2], srcPointCornersSave[3], new Scalar(255, 0, 0), 3);
Imgproc.line(SrcMat, srcPointCornersSave[3], srcPointCornersSave[0], new Scalar(255, 0, 0), 3);
SrcMat.copyTo(cameraFeed);
keypoints1.release();
img1Mat.release();
SrcMat.release();
return(true);
}
//============================================================
//=================以下為沒有再使用的函式=====================
//============================================================
//找出特徵的顏色方法三(ORB特徵點比對)
public bool descriptorsORB(Mat RGB, Mat cameraFeed, string targetName)
{
if (RGB == null)
{
Debug.Log("RGB Mat is Null");
return(false);
}
//將傳入的RGB存入Src
Mat SrcMat = new Mat();
RGB.copyTo(SrcMat);
//比對樣本載入
Texture2D imgTexture = Resources.Load(targetName) as Texture2D;
//Texture2D轉Mat
Mat targetMat = new Mat(imgTexture.height, imgTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture, targetMat);
//創建 ORB的特徵點裝置
FeatureDetector detector = FeatureDetector.create(FeatureDetector.ORB);
DescriptorExtractor extractor = DescriptorExtractor.create(DescriptorExtractor.ORB);
//產生存放特徵點Mat
MatOfKeyPoint keypointsTarget = new MatOfKeyPoint();
Mat descriptorsTarget = new Mat();
MatOfKeyPoint keypointsSrc = new MatOfKeyPoint();
Mat descriptorsSrc = new Mat();
//找特徵點圖Target
detector.detect(targetMat, keypointsTarget);
extractor.compute(targetMat, keypointsTarget, descriptorsTarget);
//找特徵點圖Src
detector.detect(SrcMat, keypointsSrc);
extractor.compute(SrcMat, keypointsSrc, descriptorsSrc);
//創建特徵點比對物件
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
MatOfDMatch matches = new MatOfDMatch();
//丟入兩影像的特徵點
matcher.match(descriptorsTarget, descriptorsSrc, matches);
DMatch[] arrayDmatch = matches.toArray();
//做篩選
double max_dist = 0;
double min_dist = 100;
//-- Quick calculation of max and min distances between keypoints
double dist = new double();
for (int i = 0; i < matches.rows(); i++)
{
dist = arrayDmatch[i].distance;
if (dist < min_dist)
{
min_dist = dist;
}
if (dist > max_dist)
{
max_dist = dist;
}
}
Debug.Log("Max dist :" + max_dist);
Debug.Log("Min dist :" + min_dist);
List <DMatch> matchesGoodList = new List <DMatch>();
MatOfDMatch matchesGood = new MatOfDMatch();
matchesGood.fromList(matchesGoodList);
//Draw Keypoints
Features2d.drawKeypoints(SrcMat, keypointsSrc, SrcMat);
List <Point> pTarget = new List <Point>();
List <Point> pSrc = new List <Point>();
Debug.Log("MatchCount" + matchesGoodList.Count);
for (int i = 0; i < matchesGoodList.Count; i++)
{
pTarget.Add(new Point(keypointsTarget.toArray()[matchesGoodList[i].queryIdx].pt.x, keypointsTarget.toArray()[matchesGoodList[i].queryIdx].pt.y));
pSrc.Add(new Point(keypointsSrc.toArray()[matchesGoodList[i].trainIdx].pt.x, keypointsSrc.toArray()[matchesGoodList[i].trainIdx].pt.y));
}
MatOfPoint2f p2fTarget = new MatOfPoint2f(pTarget.ToArray());
MatOfPoint2f p2fSrc = new MatOfPoint2f(pSrc.ToArray());
Mat matrixH = Calib3d.findHomography(p2fTarget, p2fSrc, Calib3d.RANSAC, 3);
List <Point> srcPointCorners = new List <Point>();
srcPointCorners.Add(new Point(0, 0));
srcPointCorners.Add(new Point(targetMat.width(), 0));
srcPointCorners.Add(new Point(targetMat.width(), targetMat.height()));
srcPointCorners.Add(new Point(0, targetMat.height()));
Mat originalRect = Converters.vector_Point2f_to_Mat(srcPointCorners);
List <Point> srcPointCornersEnd = new List <Point>();
srcPointCornersEnd.Add(new Point(0, targetMat.height()));
srcPointCornersEnd.Add(new Point(0, 0));
srcPointCornersEnd.Add(new Point(targetMat.width(), 0));
srcPointCornersEnd.Add(new Point(targetMat.width(), targetMat.height()));
Mat changeRect = Converters.vector_Point2f_to_Mat(srcPointCornersEnd);
Core.perspectiveTransform(originalRect, changeRect, matrixH);
List <Point> srcPointCornersSave = new List <Point>();
Converters.Mat_to_vector_Point(changeRect, srcPointCornersSave);
if ((srcPointCornersSave[2].x - srcPointCornersSave[0].x) < 5 || (srcPointCornersSave[2].y - srcPointCornersSave[0].y) < 5)
{
Debug.Log("Match Out Put image is to small");
SrcMat.copyTo(cameraFeed);
SrcMat.release();
Imgproc.putText(cameraFeed, targetName, srcPointCornersSave[0], 0, 1, new Scalar(255, 255, 255), 2);
return(false);
}
//畫出框框
Imgproc.line(SrcMat, srcPointCornersSave[0], srcPointCornersSave[1], new Scalar(255, 0, 0), 3);
Imgproc.line(SrcMat, srcPointCornersSave[1], srcPointCornersSave[2], new Scalar(255, 0, 0), 3);
Imgproc.line(SrcMat, srcPointCornersSave[2], srcPointCornersSave[3], new Scalar(255, 0, 0), 3);
Imgproc.line(SrcMat, srcPointCornersSave[3], srcPointCornersSave[0], new Scalar(255, 0, 0), 3);
//畫中心
Point middlePoint = new Point((srcPointCornersSave[0].x + srcPointCornersSave[2].x) / 2, (srcPointCornersSave[0].y + srcPointCornersSave[2].y) / 2);
Imgproc.line(SrcMat, middlePoint, middlePoint, new Scalar(0, 0, 255), 10);
SrcMat.copyTo(cameraFeed);
keypointsTarget.release();
targetMat.release();
SrcMat.release();
return(true);
}
//public bool verificaImagem(Texture2D texture, Texture2D texture2)
//{
// Texture2D camFoto = texture;
// Texture2D printTela = texture2;
// // Escala de cinza. CV_8UC1
// Mat img1Mat = new Mat(camFoto.height, camFoto.width, CvType.CV_8UC1);
// Utils.texture2DToMat(camFoto, img1Mat);
// // Escala de cinza. CV_8UC1
// Mat img2Mat = new Mat(printTela.height, printTela.width, CvType.CV_8UC1);
// Utils.texture2DToMat(printTela, img2Mat);
// Imgproc.GaussianBlur(img1Mat, img1Mat, new Size(5, 5), 0);
// Imgproc.threshold(img1Mat, img1Mat, 100, 255, Imgproc.THRESH_BINARY);
// Imgproc.GaussianBlur(img2Mat, img2Mat, new Size(5, 5), 0);
// Imgproc.threshold(img2Mat, img2Mat, 240, 255, Imgproc.THRESH_BINARY);
// //Create the result mat
// int result_cols = img1Mat.cols() - img2Mat.cols() + 1;
// int result_rows = img1Mat.rows() - img2Mat.rows() + 1;
// Mat result = new Mat(result_rows, result_cols, CvType.CV_32FC1);
// int match_method = Imgproc.TM_CCOEFF_NORMED;
// Imgproc.matchTemplate(img1Mat, img2Mat, result, match_method);
// Debug.Log(match_method);
// return match_method <= 1;
//}
public bool verificaImagem(Texture2D textParam, Texture2D textParam2)
{
var bytes = textParam.EncodeToJPG();
//File.WriteAllBytes("imagem1_tratamento.png", bytes);
//bytes = textParam2.EncodeToJPG();
//File.WriteAllBytes("imagem2_tratamento.png", bytes);
//Texture2D imgTexture = Resources.Load("circulo") as Texture2D;
Texture2D camFoto = textParam;
Texture2D printTela = textParam2;
// Escala de cinza. CV_8UC1
Mat img1Mat = new Mat(camFoto.height, camFoto.width, CvType.CV_8UC1);
Utils.texture2DToMat(camFoto, img1Mat);
// Escala de cinza. CV_8UC1
Mat img2Mat = new Mat(printTela.height, printTela.width, CvType.CV_8UC1);
Utils.texture2DToMat(printTela, img2Mat);
Imgproc.GaussianBlur(img1Mat, img1Mat, new Size(5, 5), 0);
Texture2D tex3 = new Texture2D(img1Mat.cols(), img1Mat.rows(), TextureFormat.RGBA32, false);
//Utils.matToTexture2D(img1Mat, tex3);
//bytes = tex3.EncodeToJPG();
//File.WriteAllBytes("imagem1_tratamento_gaussian.png", bytes);
Imgproc.threshold(img1Mat, img1Mat, 100, 255, Imgproc.THRESH_BINARY);
tex3 = new Texture2D(img1Mat.cols(), img1Mat.rows(), TextureFormat.RGBA32, false);
Utils.matToTexture2D(img1Mat, tex3);
bytes = tex3.EncodeToJPG();
File.WriteAllBytes("imagem1_tratamento_threshold.png", bytes);
Imgproc.GaussianBlur(img2Mat, img2Mat, new Size(5, 5), 0);
Texture2D tex4 = new Texture2D(img2Mat.cols(), img2Mat.rows(), TextureFormat.RGBA32, false);
//Utils.matToTexture2D(img2Mat, tex4);
//bytes = tex4.EncodeToJPG();
//File.WriteAllBytes("imagem2_tratamento_gaussian.png", bytes);
Imgproc.threshold(img2Mat, img2Mat, 240, 255, Imgproc.THRESH_BINARY);
tex4 = new Texture2D(img2Mat.cols(), img2Mat.rows(), TextureFormat.RGBA32, false);
Utils.matToTexture2D(img2Mat, tex4);
bytes = tex4.EncodeToJPG();
File.WriteAllBytes("imagem2_tratamento_threshold.png", bytes);
ORB detector = ORB.create();
ORB extractor = ORB.create();
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
Mat descriptors1 = new Mat();
detector.detect(img1Mat, keypoints1);
extractor.compute(img1Mat, keypoints1, descriptors1);
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
Mat descriptors2 = new Mat();
detector.detect(img2Mat, keypoints2);
extractor.compute(img2Mat, keypoints2, descriptors2);
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
MatOfDMatch matches = new MatOfDMatch();
matcher.match(descriptors1, descriptors2, matches);
List <MatOfDMatch> lista = new List <MatOfDMatch>();
lista.Add(matches);
matcher.knnMatch(descriptors1, descriptors2, lista, 2);
long total = 0;
foreach (MatOfDMatch item in lista)
{
if (item.toList()[0].distance < 0.75 * item.toList()[1].distance)
{
total++;
}
}
long number_keypoints = 0;
if (keypoints1.elemSize() <= keypoints2.elemSize())
{
number_keypoints = keypoints1.elemSize();
}
else
{
number_keypoints = keypoints2.elemSize();
}
Debug.Log(total / number_keypoints * 100);
return((total / number_keypoints * 100) >= 70);
}
void Orb()
{
p1Mat = Imgcodecs.imread(Application.dataPath + "/Textures/1.jpg", 1);
p2Mat = Imgcodecs.imread(Application.dataPath + "/Textures/3.jpg", 1);
Imgproc.cvtColor(p1Mat, p1Mat, Imgproc.COLOR_BGR2RGB);
Imgproc.cvtColor(p2Mat, p2Mat, Imgproc.COLOR_BGR2RGB);
Imgproc.resize(p2Mat, p2Mat, new Size(p1Mat.width(), p1Mat.height()));
Debug.Log(p2Mat);
/*
* //仿射变换（矩阵旋转）
* float angle = UnityEngine.Random.Range(0, 360), scale = 1.0f;
* Point center = new Point(img2Mat.cols() * 0.5f, img2Mat.rows() * 0.5f);
*
* Mat affine_matrix = Imgproc.getRotationMatrix2D(center, angle, scale);
* Imgproc.warpAffine(img1Mat, img2Mat, affine_matrix, img2Mat.size());
*
* Texture2D texture = new Texture2D(img2Mat.cols(), img2Mat.rows());
* Utils.matToTexture2D(img2Mat, texture);
* outputRawImage.texture = texture;
*/
ORB detector = ORB.create();
ORB extractor = ORB.create();
//提取图一特征点
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
Mat descriptors1 = new Mat();
detector.detect(p1Mat, keypoints1);
extractor.compute(p1Mat, keypoints1, descriptors1);
//提取图二特征点
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
Mat descriptors2 = new Mat();
detector.detect(p2Mat, keypoints2);
extractor.compute(p2Mat, keypoints2, descriptors2);
//第一次匹配结果（密密麻麻）
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
MatOfDMatch matches = new MatOfDMatch();
matcher.match(descriptors1, descriptors2, matches);
//筛选（非官方）
//计算向量距离的最大值/最小值
double max_dist = 0;
double min_dist = 15; //通过距离控制需要的特征。
//（设到10，最终只有2个耳朵匹配。。。）
//（设到15，尾巴也开始匹配。。。。。。）
//新建两个容器存放筛选样本
List <DMatch> matchesArray = matches.toList(); //用Unity版API多转一步
//Debug.Log(matchesArray.Count); //500
List <DMatch> goodmatchesArray = new List <DMatch>();
//Debug.Log(img1Mat.rows()); //512
for (int i = 0; i < matchesArray.Count; i++)
{
Debug.Log("[" + i + "]" + matchesArray[i].distance);
if (matchesArray[i].distance > max_dist)
{
//max_dist = matchesArray[i].distance;
}
if (matchesArray[i].distance < min_dist)
{
min_dist = matchesArray[i].distance;
}
}
//Debug.Log("The max distance is: " + max_dist);
Debug.Log("The min distance is: " + min_dist);
for (int i = 0; i < matchesArray.Count; i++)
{
if (matchesArray[i].distance < 2 * min_dist) //
{
goodmatchesArray.Add(matchesArray[i]);
}
}
MatOfDMatch newMatches = new MatOfDMatch();
newMatches.fromList(goodmatchesArray);
Debug.Log(newMatches.toList().Count); //第二次筛选后符合的
//绘制第二次筛选结果
dstMat = new Mat();
Features2d.drawMatches(p1Mat, keypoints1, p2Mat, keypoints2, newMatches, dstMat);
Texture2D t2d = new Texture2D(dstMat.width(), dstMat.height());
Utils.matToTexture2D(dstMat, t2d);
Sprite sp = Sprite.Create(t2d, new UnityEngine.Rect(0, 0, t2d.width, t2d.height), Vector2.zero);
m_dstImage.sprite = sp;
m_dstImage.preserveAspect = true;
}
public ImageString MatchFeatures(string base64image, List <string> base64imageList)
{
List <MatOfDMatch> winnerMatches = new List <MatOfDMatch>();
MatOfKeyPoint winnerKeyPoints = new MatOfKeyPoint();
Mat winnerImage = new Mat();
int winnerIndex = -1;
int winnerValue = 0;
Texture2D imgTexture = base64ImageToTexture(base64image);
List <Texture2D> imgTextures = new List <Texture2D>();
for (int i = 0; i < base64imageList.Count; i++)
{
imgTextures.Add(base64ImageToTexture(base64imageList[i]));
}
//Create Mat from texture
Mat img1Mat = new Mat(imgTexture.height, imgTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture, img1Mat);
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
Mat descriptors1 = new Mat();
FeatureDetector detector = FeatureDetector.create(FeatureDetector.ORB);
DescriptorExtractor extractor = DescriptorExtractor.create(DescriptorExtractor.ORB);
//Detect keypoints and compute descriptors from photo.
detector.detect(img1Mat, keypoints1);
extractor.compute(img1Mat, keypoints1, descriptors1);
Debug.Log("Billede features: " + descriptors1.rows());
if (descriptors1.rows() < 10)
{
Debug.Log("ARRRRRRGH der er ikke mange descripters i mit original-billede");
return(new ImageString(base64image, winnerIndex));
}
//Run through each image in list
for (int i = 0; i < imgTextures.Count; i++)
{
Texture2D imgTexture2 = imgTextures[i];
//Create Mat from texture
Mat img2Mat = new Mat(imgTexture2.height, imgTexture2.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture2, img2Mat);
//Find keypoints and descriptors from image in list
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
Mat descriptors2 = new Mat();
detector.detect(img2Mat, keypoints2);
extractor.compute(img2Mat, keypoints2, descriptors2);
//Match photo with image from list
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
Debug.Log("Billede2 features: " + descriptors2.rows());
if (descriptors2.rows() < 10)
{
Debug.Log("ARRRRRRGH der er ikke mange descripters i mit test billede: " + i);
continue;
}
List <MatOfDMatch> matchList = new List <MatOfDMatch>();
matcher.knnMatch(descriptors1, descriptors2, matchList, 2);
//Find the good matches and put them ind a list
List <MatOfDMatch> good = new List <MatOfDMatch>();
foreach (MatOfDMatch match in matchList)
{
DMatch[] arrayDmatch = match.toArray();
if (arrayDmatch[0].distance < 0.7f * arrayDmatch[1].distance)
{
good.Add(match);
}
}
//Find the best match image based on the good lists
if (good.Count > winnerThreshold && good.Count > winnerValue)
{
winnerImage = img2Mat;
winnerMatches = good;
winnerKeyPoints = keypoints2;
winnerIndex = i;
winnerValue = good.Count;
}
}
Debug.Log("The winner is image: " + winnerIndex + " with a value of: " + winnerValue);
//If no winner just return the original image
if (winnerIndex == -1)
{
Debug.Log("No winner");
return(new ImageString(base64image, winnerIndex));
}
Debug.Log("No winner");
//Find the matching keypoints from the winner list.
MatOfPoint2f queryPoints = new MatOfPoint2f();
MatOfPoint2f matchPoints = new MatOfPoint2f();
List <Point> queryPointsList = new List <Point>();
List <Point> matchPointsList = new List <Point>();
foreach (MatOfDMatch match in winnerMatches)
{
DMatch[] arrayDmatch = match.toArray();
queryPointsList.Add(keypoints1.toList()[arrayDmatch[0].queryIdx].pt);
matchPointsList.Add(winnerKeyPoints.toList()[arrayDmatch[0].trainIdx].pt);
}
queryPoints.fromList(queryPointsList);
matchPoints.fromList(matchPointsList);
//Calculate the homography of the best matching image
Mat homography = Calib3d.findHomography(queryPoints, matchPoints, Calib3d.RANSAC, 5.0);
Mat resultImg = new Mat();
Imgproc.warpPerspective(img1Mat, resultImg, homography, winnerImage.size());
//Show image
Texture2D texture = new Texture2D(winnerImage.cols(), winnerImage.rows(), TextureFormat.RGBA32, false);
Utils.matToTexture2D(resultImg, texture);
return(new ImageString(Convert.ToBase64String(texture.EncodeToPNG()), winnerIndex));
}
public List <ImageObject> MatchFeatures(string base64image, List <string> base64imageList)
{
ImageObject myImage = new ImageObject();
ImageObject winnerImage = new ImageObject();
List <ImageObject> returnImageList = new List <ImageObject>();
Texture2D imgTexture = base64ImageToTexture(base64image);
List <Texture2D> imgTextures = new List <Texture2D>();
for (int i = 0; i < base64imageList.Count; i++)
{
imgTextures.Add(base64ImageToTexture(base64imageList[i]));
}
//Create Mat from texture
Mat img1Mat = new Mat(imgTexture.height, imgTexture.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture, img1Mat);
MatOfKeyPoint keypoints1 = new MatOfKeyPoint();
Mat descriptors1 = new Mat();
FeatureDetector detector = FeatureDetector.create(FeatureDetector.ORB);
DescriptorExtractor extractor = DescriptorExtractor.create(DescriptorExtractor.ORB);
//Detect keypoints and compute descriptors from photo.
detector.detect(img1Mat, keypoints1);
extractor.compute(img1Mat, keypoints1, descriptors1);
//Debug.Log("Billede features: " + descriptors1.rows());
myImage.image = base64image;
myImage.keyPoints = keypoints1;
myImage.imageMat = img1Mat;
if (descriptors1.rows() < 10)
{
Debug.Log("ARRRRRRGH der er ikke mange descripters i mit original-billede");
//No winner as there is to few descriptors.
return(returnImageList);
}
//Run through each image in list
//-------------------------------------------------------------
for (int i = 0; i < imgTextures.Count; i++)
{
Texture2D imgTexture2 = imgTextures[i];
//Create Mat from texture
Mat img2Mat = new Mat(imgTexture2.height, imgTexture2.width, CvType.CV_8UC3);
Utils.texture2DToMat(imgTexture2, img2Mat);
//Find keypoints and descriptors from image in list
MatOfKeyPoint keypoints2 = new MatOfKeyPoint();
Mat descriptors2 = new Mat();
detector.detect(img2Mat, keypoints2);
extractor.compute(img2Mat, keypoints2, descriptors2);
//Match photo with image from list
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMINGLUT);
//Debug.Log("Billede2 features: " + descriptors2.rows());
if (descriptors2.rows() < 10)
{
Debug.Log("ARRRRRRGH der er ikke mange descripters i mit test billede: " + i);
continue;
}
List <MatOfDMatch> matchList = new List <MatOfDMatch>();
matcher.knnMatch(descriptors1, descriptors2, matchList, 2);
//Find the good matches and put them ind a list
List <MatOfDMatch> good = new List <MatOfDMatch>();
foreach (MatOfDMatch match in matchList)
{
DMatch[] arrayDmatch = match.toArray();
if (arrayDmatch[0].distance < 0.7f * arrayDmatch[1].distance)
{
good.Add(match);
}
}
//Find the best match image based on the good lists
if (good.Count > winnerThreshold && good.Count > winnerImage.value)
{
winnerImage.index = i;
winnerImage.imageMat = img2Mat;
winnerImage.keyPoints = keypoints2;
winnerImage.value = good.Count;
winnerImage.matches = good;
}
}
// Run through done
//-------------------------------------------------------------
Debug.Log("The winner is image: " + winnerImage.index + " with a value of: " + winnerImage.value);
//If no winner just return the original image
if (winnerImage.index == -1)
{
Debug.Log("No winner");
return(returnImageList);
}
Texture2D imageTexture = new Texture2D(winnerImage.imageMat.cols(), winnerImage.imageMat.rows(), TextureFormat.RGBA32, false);
winnerImage.image = Convert.ToBase64String(imageTexture.EncodeToPNG());
returnImageList.Add(myImage);
returnImageList.Add(winnerImage);
return(returnImageList);
}