up

CMakeLists.txt

@@ -3,12 +3,6 @@ project(stitch VERSION 0.1.0 LANGUAGES C CXX)
 
 SET(ArithStitchDir stitch)
 
-IF(WIN32)
-    set(OpenCV_DIR "C:/opencv/build/x64/vc14/lib")
-    set(CMAKE_TOOLCHAIN_FILE "C:/Users/75929/vcpkg/scripts/buildsystems/vcpkg.cmake")
-ELSE(WIN32)
-    set(OpenCV_DIR "/home/wcw/opencv-3.4.16/install/share/OpenCV")
-ENDIF(WIN32)                
 
 
 find_package(OpenCV REQUIRED)

main.cpp

@@ -94,17 +94,18 @@ int main(int, char**)
     cv::Mat mat_pan;//全景显示
     //return 0;
     FILE* file = fopen("../20241219152917_4.xraw","rb");
+    //FILE* file = fopen("/home/wang/data62/729dataset/ir_1280_1024_para40_y16/5.xraw","rb");
   
     for (size_t i = 0; i < 900; i++)
     {
 
         fread(pFrameIR,2,IMAGE_WIDTH_IR*(IMAGE_HEIGHT_IR + PARA_IR_LINE),file);
 
-        
-
         S729paras_IR Paras_IR = { 0 };
 	    memcpy(&Paras_IR, (unsigned char*)(pFrameIR + IMAGE_WIDTH_IR*IMAGE_HEIGHT_IR), sizeof(S729paras_IR));
 
+
+        
         FrameInfo info = {0};
         info.nFrmID = i;
         info.camInfo.nFocus = Paras_IR.Paras_IR.caminfo.nFocal;
@@ -132,13 +133,6 @@ int main(int, char**)
         cv::Mat mat_src(IMAGE_HEIGHT_IR, IMAGE_WIDTH_IR, CV_8UC1, pImageIR);
 
 
-        //auto  detector_ = cv::ORB::create();
-        //std::vector<cv::KeyPoint> keypoints; 
-        //cv::Mat descriptors;
-        //detector_->detectAndCompute(mat_src, cv::noArray(), keypoints, descriptors);
-
-
-
 		frame.enPixelFormat = GD_PIXEL_FORMAT_GRAY_Y8;
 		frame.u32Width = IMAGE_WIDTH_IR;
 		frame.u32Height = IMAGE_HEIGHT_IR;
@@ -158,23 +152,39 @@ int main(int, char**)
         }
         else
         {
-            //cout << info.craft.stPos.B << " " <<info.craft.stPos.L << " " << info.craft.stPos.H << " "
+            if(i % 20 != 0)
-            //<< info.craft.stAtt.fYaw << " " << info.craft.stAtt.fPitch << " " << info.craft.stAtt.fRoll<< " "
+            {
-            //<< info.servoInfo.fServoAz << " " << info.servoInfo.fServoPt
+                continue;
-            //<< endl;
+            }
+            cout << info.craft.stPos.B << " " <<info.craft.stPos.L << " " << info.craft.stPos.H << " "
+            << info.craft.stAtt.fYaw << " " << info.craft.stAtt.fPitch << " " << info.craft.stAtt.fRoll<< " "
+            << info.servoInfo.fServoAz << " " << info.servoInfo.fServoPt
+            << endl;
 
+            // 基于地理系的拼接
+            //stitcher->GeoStitch(frame,info);
 
-            stitcher->GeoStitch(frame,info);
+            // 基于地理+BA拼接
+            auto nNum = stitcher->BAStitch(frame,info);
 
+            //cout << nNum << endl;
 
+            //imshow("src", mat_src);
+            //waitKey(0);
         }
-        //imshow("src", mat_src);
+
-        imshow("pan", mat_pan);
+        // imshow("pan", mat_pan);
-        waitKey(1);
+        // waitKey(1);
+
+
 
     }
 
 
-    waitKey(0);
+    // 读入所有帧，执行一次全局BA
+    stitcher->BAOpt();
+
+
+    //waitKey(0);
 
 }

stitch/src/API_VideoStitch.h

@@ -28,8 +28,10 @@ public:
     // 几何校正快拼
     virtual BYTE8       GeoStitch(GD_VIDEO_FRAME_S img, FrameInfo para) = 0;
 
-    // 基于特征点拼接，全局光束平差方法
+    // 基于特征点拼接，全局光束平差方法，分为更新和优化调整两个接口
-    virtual BYTE8       BAStitch(GD_VIDEO_FRAME_S img, FrameInfo para) = 0;
+    virtual SINT32       BAStitch(GD_VIDEO_FRAME_S img, FrameInfo para) = 0;
+    virtual SINT32       BAOpt() = 0;
+
 
 	// 获取全景图
 	virtual GD_VIDEO_FRAME_S ExportPanAddr() = 0;

stitch/src/Arith_BATask.cpp

@@ -10,9 +10,12 @@ struct CostFunctor {
     }
 };
 
-BA_Task::BA_Task()
+
+
+BA_Task::BA_Task(GeoStitcher * pGeoTrans)
 {
-    // 来个100帧
+    _GeoStitcher = pGeoTrans;
+    _FeaMatcher = new FeatureMatcher(DetectorType::ORB, MatcherType::BF);
     _imgVec.reserve(100);
 }
 
@@ -22,17 +25,44 @@ BA_Task::~BA_Task()
 
 SINT32 BA_Task::addFrame(GD_VIDEO_FRAME_S img, FrameInfo para)
 {
+    cv::Mat frame = Mat(img.u32Height, img.u32Width, CV_8UC1, (void*)img.u64VirAddr[0]);
+
     // 缓存图像帧
     if (img.enPixelFormat == GD_PIXEL_FORMAT_E::GD_PIXEL_FORMAT_GRAY_Y8)
     {
-        _imgVec.emplace_back(Mat(img.u32Height, img.u32Width, CV_8UC1, img.u64VirAddr[0]));
+        _imgVec.emplace_back(frame.clone());
     }
+    // 缓存外参
+    _paraVec.push_back(para);
+
+    // 缓存初始H
+    Proj t_Proj = _GeoStitcher->AnlayseTform(para);
+    auto H = _GeoStitcher->findHomography(t_Proj,_panPara);
+    _origMatrix.push_back(H);
+    _currMatrix.push_back(H);
+
+
+    // 提取特征点
+    std::vector<cv::KeyPoint> keypoints;
+    cv::Mat descriptors;
+    _FeaMatcher->extractFeatures(frame,keypoints,descriptors);
+    _FeaPtVec.push_back(keypoints);
+    _FeaDespVec.push_back(descriptors);
+
+    // 缓存包围多边形
+    _polygon.push_back(warpRectWithH(H,cv::Size(img.u32Width,img.u32Height)));
+
 
 
 
     return _imgVec.size();
 }
 
+void BA_Task::setPanPara(PanInfo info)
+{
+    _panPara = info;
+}
+
 
 
 
@@ -40,28 +70,151 @@ SINT32 BA_Task::addFrame(GD_VIDEO_FRAME_S img, FrameInfo para)
 
 void BA_Task::Test()
 {
-    google::InitGoogleLogging("ceres");
+    // google::InitGoogleLogging("ceres");
-
+
-    // The variable to solve for with its initial value. It will be
+    // // The variable to solve for with its initial value. It will be
-    // mutated in place by the solver.
+    // // mutated in place by the solver.
-    double x = 0.5;
+    // double x = 0.5;
-    const double initial_x = x;
+    // const double initial_x = x;
-
+
-    // Build the problem.
+    // // Build the problem.
-    ceres::Problem problem;
+    // ceres::Problem problem;
-
+
-    // Set up the only cost function (also known as residual). This uses
+    // // Set up the only cost function (also known as residual). This uses
-    // auto-differentiation to obtain the derivative (jacobian).
+    // // auto-differentiation to obtain the derivative (jacobian).
-    ceres::CostFunction* cost_function =
+    // ceres::CostFunction* cost_function =
-        new ceres::AutoDiffCostFunction<CostFunctor, 1, 1>(new CostFunctor);
+    //     new ceres::AutoDiffCostFunction<CostFunctor, 1, 1>(new CostFunctor);
-    problem.AddResidualBlock(cost_function, nullptr, &x);
+    // problem.AddResidualBlock(cost_function, nullptr, &x);
-
+
-    // Run the solver!
+    // // Run the solver!
-    ceres::Solver::Options options;
+    // ceres::Solver::Options options;
-    options.minimizer_progress_to_stdout = true;
+    // options.minimizer_progress_to_stdout = true;
-    ceres::Solver::Summary summary;
+    // ceres::Solver::Summary summary;
-    ceres::Solve(options, &problem, &summary);
+    // ceres::Solve(options, &problem, &summary);
-
+
-    std::cout << summary.BriefReport() << "\n";
+    // std::cout << summary.BriefReport() << "\n";
-    std::cout << "x : " << initial_x << " -> " << x << "\n";
+    // std::cout << "x : " << initial_x << " -> " << x << "\n";
+
+    
+}
+
+void BA_Task::optimizeBA()
+{
+    // 计算匹配性矩阵
+    CalMatchMat(0.3);
+
+
+}
+
+SINT32 BA_Task::CalMatchMat(float fiou_thre)
+{
+    _IOUMat = cv::Mat::zeros(_polygon.size(),_polygon.size(),CV_32FC1);
+    // 先计算IOU矩阵
+    for (size_t i = 0; i < _polygon.size(); i++)
+    {
+        vector<cv::Point2f> poly_i = _polygon[i];
+
+        for (size_t j = 0; j < _polygon.size(); j++)
+        {
+            if (i == j)
+            {
+                _IOUMat.at<float>(i,j) = 1; 
+                continue;
+            }
+            vector<cv::Point2f> poly_j = _polygon[j];
+
+            float fiou = computeQuadrilateralIOU(poly_i,poly_j);
+            _IOUMat.at<float>(i,j) = fiou;
+            _IOUMat.at<float>(j,i) = fiou;//是对称矩阵
+            
+        }
+    }
+
+    // IOU大于0.3以上才进行特征点匹配
+    for (size_t i = 0; i < _polygon.size(); i++)
+    {
+        for (size_t j = 0; j < _polygon.size(); j++)
+        {
+            if (i == j)
+            {
+                continue;
+            }
+
+            if (_IOUMat.at<float>(i,j) < fiou_thre)
+            {
+                continue;
+            }
+
+
+            // IOU满足条件才匹配特征点
+            cv::Mat image(1000, 1000, CV_8UC3, cv::Scalar(255, 255, 255));
+            vector<vector<cv::Point2f>> tmpPoly;
+            tmpPoly.push_back(_polygon[i]);
+            tmpPoly.push_back(_polygon[j]);
+            drawPolygons(image, tmpPoly);
+
+            // 显示绘制结果
+            cv::imshow("Polygons", image);
+            cv::waitKey(1);
+            std::vector<cv::DMatch> matches;
+            _FeaMatcher->matchFeatures(_FeaDespVec[i],_FeaDespVec[j],matches);
+
+            // 可视化匹配结果
+            cv::Mat img_matches;
+            cv::drawMatches(
+                _imgVec[i], _FeaPtVec[i], // 第一幅图像及其特征点
+                _imgVec[j], _FeaPtVec[j], // 第二幅图像及其特征点
+                matches,          // 匹配结果
+                img_matches,      // 输出图像
+                cv::Scalar::all(-1), // 匹配线颜色（默认随机颜色）
+                cv::Scalar::all(-1), // 未匹配点颜色（默认不绘制）
+                std::vector<char>(), // 掩码（可选，用于筛选匹配）
+                cv::DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS // 不绘制未匹配的点
+            );
+
+            cv::resize(img_matches,img_matches,cv::Size(1280,512));
+
+            // 显示匹配结果
+            cv::imshow("Feature Matches", img_matches);
+            cv::waitKey(0);
+
+        }
+    }
+    
+
+
+
+    return 0;
+}
+
+
+
+
+// 函数：绘制多边形
+void drawPolygons(cv::Mat& image, const std::vector<std::vector<cv::Point2f>>& polygons) 
+{
+    // 定义颜色列表
+    std::vector<cv::Scalar> colors = {
+        cv::Scalar(255, 0, 0),   // 蓝色
+        cv::Scalar(0, 255, 0),   // 绿色
+        cv::Scalar(0, 0, 255),   // 红色
+        cv::Scalar(255, 255, 0), // 青色
+        cv::Scalar(255, 0, 255), // 品红
+        cv::Scalar(0, 255, 255)  // 黄色
+    };
+
+    // 绘制每个多边形
+    for (size_t i = 0; i < polygons.size(); ++i) {
+        const auto& polygon = polygons[i];
+        cv::Scalar color = colors[i % colors.size()]; // 循环使用颜色
+
+        // 将多边形点转换为整数类型
+        std::vector<cv::Point> intPolygon;
+        for (const auto& pt : polygon) {
+            intPolygon.push_back(cv::Point(static_cast<int>(pt.x), static_cast<int>(pt.y)));
+        }
+
+        // 绘制多边形
+        cv::polylines(image, intPolygon, true, color, 2);
+    }
 }

stitch/src/Arith_BATask.h

@@ -11,7 +11,10 @@
 *****************************************************************************************/
 #pragma once
 #include "StitchStruct.h"
-
+#include <opencv2/core.hpp>
+#include <opencv2/imgproc.hpp>
+#include "Arith_GeoStitcher.h"
+#include "Arith_FeaMatch.h"
 using cv::Mat_;
 
 
@@ -21,24 +24,47 @@ using cv::Mat_;
 class BA_Task
 {
 public:
-	BA_Task();
+	BA_Task(GeoStitcher* pGeoTrans);
 	~BA_Task();
 
+	// 加入优化帧
 	SINT32 addFrame(GD_VIDEO_FRAME_S img, FrameInfo para);
 
+	// 设置全景参数
+	void setPanPara(PanInfo info);
+
 	void Test();
+
+	// BA优化
+	void optimizeBA();
+
 private:
-	Mat_<int> _similarityMat;//配准点邻接表
+	SINT32 CalMatchMat(float fiou_thre);//计算匹配性矩阵(以IOU约束)
 
+	SINT32 GuideMatch(int i,int j);//外参指示下的匹配
 
+private:
+	GeoStitcher* _GeoStitcher;//外参计算
+	FeatureMatcher* _FeaMatcher;//特征匹配
+	PanInfo _panPara;//全景图配置
 
-	//超大空间，考虑文件cache todo！
 private:
-	vector<cv::Mat> _imgVec;//图像缓存
+	Mat_<int> _MatchMat;//配准点邻接表
-	vector<cv::Point2d> _FeaPtVec;//特征点缓存
+	Mat_<float> _IOUMat;//交汇邻接表
 
+	vector<cv::Mat> _origMatrix;//初始H矩阵
+	vector<cv::Mat> _currMatrix;//当前H矩阵
 
+	//超大空间，考虑文件cache todo！
+private:
+	vector<cv::Mat> _imgVec;//图像缓存
+	vector<vector<cv::KeyPoint>> _FeaPtVec;//特征点缓存
+	vector<cv::Mat> _FeaDespVec;//特征点描述子
+	vector<FrameInfo> _paraVec;
+	vector<vector<cv::Point2f>> _polygon;//帧包围四边形
 
 };
 	
 
+
+void drawPolygons(cv::Mat& image, const std::vector<std::vector<cv::Point2f>>& polygons);

stitch/src/Arith_FeaMatch.cpp

@@ -57,7 +57,7 @@ void FeatureMatcher::initDetector()
 #endif
         break;
     case ORB:
-        detector_ = cv::ORB::create();
+        detector_ = cv::ORB::create(1000);
         break;
     default:
         throw std::invalid_argument("Unsupported feature detector type");

stitch/src/Arith_FeaMatch.h

@@ -26,7 +26,6 @@ class FeatureMatcher
 {
 public:
     
-
     // 构造函数
     FeatureMatcher(DetectorType detectorType, MatcherType matcherType);
 

stitch/src/Arith_GeoStitcher.cpp

@@ -28,14 +28,14 @@ PanInfo GeoStitcher::InitMap(FrameInfo info)
 	PanInfo panPara = {0};
 	panPara.m_pan_width = 1000;//全景宽
 	panPara.m_pan_height = 1000;//全景高
-	panPara.scale = 0.5;//比例尺,1m = ?pix
+	panPara.scale = 0.3;//比例尺,1m = ?pix
 
 	// 直接无平移解算
 	auto cur = Trans_Geo2pan(ct_geo, panPara);
 
 	// 计算平移到全景图固定点的平移量，从此处开始拼接
-	int planX = 200;
+	int planX = 500;
-	int planY = 100;
+	int planY = 500;
 
 	panPara.map_shiftX = planX - (cur.x);//平移X
 	panPara.map_shiftY = planY - (cur.y);//平移Y
@@ -212,6 +212,35 @@ Proj GeoStitcher::AnlayseTform(FrameInfo info)
 	return projection;
 }
 
+cv::Mat GeoStitcher::findHomography(Proj proj, PanInfo pan)
+{
+	// 同名点计算，从像方到全景
+	cv::Point2f leftTop_map = back_project(cv::Point2f(0,0), proj, pan);
+	cv::Point2f rightTop_map = back_project(cv::Point2f(1000,0), proj, pan);
+	cv::Point2f rightBottom_map = back_project(cv::Point2f(1000,1000), proj, pan);
+	cv::Point2f leftBottom_map = back_project(cv::Point2f(0,1000), proj, pan);
+
+
+    std::vector<cv::Point2f> srcPoints;
+    srcPoints.push_back(cv::Point2f(0,0)); 
+    srcPoints.push_back(cv::Point2f(1000,0)); 
+    srcPoints.push_back(cv::Point2f(1000,1000)); 
+    srcPoints.push_back(cv::Point2f(0,1000));
+
+    // 目标图像（全景图）的四个顶点坐标
+    std::vector<cv::Point2f> dstPoints;
+    dstPoints.push_back(leftTop_map); // 左
+    dstPoints.push_back(rightTop_map); // 右上
+    dstPoints.push_back(rightBottom_map); // 右下
+    dstPoints.push_back(leftBottom_map); // 左下
+
+    // 计算单应性矩阵 H
+    cv::Mat H = cv::findHomography(srcPoints, dstPoints);
+
+
+	return H;
+}
+
 
 // 计算多边形的面积
 double polygonArea(const vector<cv::Point2f>& points)
@@ -246,3 +275,24 @@ double computeQuadrilateralIOU(const vector<cv::Point2f>& quad1, const vector<cv
 	if (unionArea == 0) return 0.0; // 避免除零错误
 	return intersectionArea / unionArea;
 }
+
+cv::Point2f warpPointWithH(cv::Mat H, cv::Point2f srcPt)
+{
+	Mat point = (Mat_<double>(3, 1) << srcPt.x, srcPt.y, 1);
+	Mat result = H * point;
+	// 转局部地理系
+	double warpedX = result.at<double>(0, 0) / result.at<double>(2, 0);
+	double warpedY = result.at<double>(1, 0) / result.at<double>(2, 0);
+
+	return cv::Point2f(warpedX,warpedY);
+}
+
+vector<cv::Point2f> warpRectWithH(cv::Mat H,cv::Size size)
+{
+	vector<cv::Point2f> _res;
+	_res.push_back(warpPointWithH(H, cv::Point2f(0,0)));
+	_res.push_back(warpPointWithH(H, cv::Point2f(size.width,0)));
+	_res.push_back(warpPointWithH(H, cv::Point2f(size.width,size.height)));
+	_res.push_back(warpPointWithH(H, cv::Point2f(0,size.height)));
+	return _res;
+}

stitch/src/Arith_GeoStitcher.h

@@ -1,6 +1,6 @@
 
 /*********版权所有（C）2025，武汉高德红外股份有限公司***************************************
-* 文件名称： Arith_ExtrinsicStitcher.h
+* 文件名称： Arith_GeoStitcher.h
 * 文件标识：吊舱视频帧外参拼接
 * 内容摘要：基于机载下视外部参数的几何校正,支持独立工作,也可以作为特征拼接的外参初始值
 * 其它说明：
@@ -51,6 +51,9 @@ public:
     // 计算当前帧像方-地理坐标系R t（反投影关系）
     Proj AnlayseTform(FrameInfo info);
 
+    // 建立H矩阵,经验证与上述反投影过程完全等效，该过程应该可以通过解析直接获得，此处不再深入研究
+    cv::Mat findHomography(Proj proj,PanInfo pan);
+
 public:
 
 
@@ -85,3 +88,9 @@ double polygonArea(const vector<cv::Point2f>& points);
 
 // 计算两个四边形的 IOU
 double computeQuadrilateralIOU(const vector<cv::Point2f>& quad1, const vector<cv::Point2f>& quad2);
+
+// H映射点
+cv::Point2f warpPointWithH(cv::Mat H,cv::Point2f srcPt);
+
+// H映射多边形，按照顺时针
+vector<cv::Point2f> warpRectWithH(cv::Mat H,cv::Size size);

stitch/src/Arith_VideoStitch.cpp

@@ -23,9 +23,9 @@ void API_VideoStitch::Destroy(API_VideoStitch * obj)
 VideoStitch::VideoStitch(SINT32 nWidth, SINT32 nHeight)
 {
 	_GeoStitcher = new GeoStitcher();
-
-	_FeaMatcher = new FeatureMatcher(DetectorType::ORB, MatcherType::FLANN);
 	memset(&_pan, 0, sizeof(GD_VIDEO_FRAME_S));
+
+	_BATask = new BA_Task(_GeoStitcher);
 }
 
 VideoStitch::~VideoStitch()
@@ -43,6 +43,8 @@ PanInfo VideoStitch::Init(FrameInfo info)
 {
 	_panPara = _GeoStitcher->InitMap(info);
 
+	_BATask->setPanPara(_panPara);
+
 	_pan = cv::Mat::zeros(_panPara.m_pan_height, _panPara.m_pan_width, CV_8UC1);
 
 	return _panPara;
@@ -59,6 +61,12 @@ BYTE8 VideoStitch::GeoStitch(GD_VIDEO_FRAME_S img, FrameInfo para)
 	cv::Point2f leftBottom_map = _GeoStitcher->back_project(cv::Point2f(0,img.u32Height), t_Proj, _panPara);
 
 
+	auto H = _GeoStitcher->findHomography(t_Proj,_panPara);
+	leftTop_map = warpPointWithH(H,cv::Point2f(0,0));
+	rightTop_map = warpPointWithH(H,cv::Point2f(img.u32Width,0));
+	rightBottom_map = warpPointWithH(H,cv::Point2f(img.u32Width,img.u32Height));
+	leftBottom_map = warpPointWithH(H,cv::Point2f(0,img.u32Height));
+
 	// 计算全景图的范围
 	int right = max(max(max(leftTop_map.x, leftBottom_map.x), rightTop_map.x), rightBottom_map.x);
 	int left = min(min(min(leftTop_map.x, leftBottom_map.x), rightTop_map.x), rightBottom_map.x);
@@ -83,6 +91,7 @@ BYTE8 VideoStitch::GeoStitch(GD_VIDEO_FRAME_S img, FrameInfo para)
 			//转换为pixel坐标
 			cv::Point2f p_img = _GeoStitcher->project(Point2f(j, i), t_Proj, _panPara);
 
+			p_img = warpPointWithH(H.inv(),Point2f(j, i));
 			if (p_img.x >= 0 && p_img.y >= 0 && p_img.x < img.u32Width && p_img.y < img.u32Height)
 			{
 				_pan.data[i * _pan.rows + j] =
@@ -95,10 +104,19 @@ BYTE8 VideoStitch::GeoStitch(GD_VIDEO_FRAME_S img, FrameInfo para)
     return 0;
 }
 
-BYTE8 VideoStitch::BAStitch(GD_VIDEO_FRAME_S img, FrameInfo para)
+SINT32 VideoStitch::BAStitch(GD_VIDEO_FRAME_S img, FrameInfo para)
 {
 	
-	return BYTE8();
+	SINT32 nFrameCnt = _BATask->addFrame(img,para);
+	
+
+	return 0;
+}
+
+SINT32 VideoStitch::BAOpt()
+{
+	_BATask->optimizeBA();
+	return 0;
 }
 
 GD_VIDEO_FRAME_S VideoStitch::ExportPanAddr()

stitch/src/Arith_VideoStitch.h

@@ -3,6 +3,7 @@
 #include "Arith_GeoStitcher.h"
 #include "Arith_FeaMatch.h"
 #include "StitchStruct.h"
+#include "Arith_BATask.h"
 
 class VideoStitch:public API_VideoStitch 
 {
@@ -18,7 +19,8 @@ public:
 
     BYTE8 GeoStitch(GD_VIDEO_FRAME_S img, FrameInfo para);
 
-    BYTE8 BAStitch(GD_VIDEO_FRAME_S img, FrameInfo para);
+    SINT32 BAStitch(GD_VIDEO_FRAME_S img, FrameInfo para);
+    SINT32 BAOpt();
 
     GD_VIDEO_FRAME_S     ExportPanAddr();
 private:
@@ -29,10 +31,9 @@ private:
 
 	GeoStitcher* _GeoStitcher;//外参计算
 
-    FeatureMatcher* _FeaMatcher;//特征匹配
-
     
 
+    BA_Task* _BATask;//BA
 
     PanInfo _panPara;//全景图配置
     cv::Mat _pan;