Any n image splicing _ effect very good _ computer Vision Big Job 1 Final edition

#include <iostream> #include <fstream> #include <string> #include "opencv2/opencv_modules.hpp" # Include "Opencv2/highgui/highgui.hpp" #include "opencv2/stitching/detail/autocalib.hpp" #include "opencv2/stitching /detail/blenders.hpp "#include" opencv2/stitching/detail/camera.hpp "#include" opencv2/stitching/detail/exposure_ Compensate.hpp "#include" opencv2/stitching/detail/matchers.hpp "#include" opencv2/stitching/detail/motion_ Estimators.hpp "#include" opencv2/stitching/detail/seam_finders.hpp "#include" opencv2/stitching/detail/util.hpp "# Include "Opencv2/stitching/detail/warpers.hpp" #include "opencv2/stitching/warpers.hpp" #include <time.h> Using namespace Std;using namespace cv;using namespace CV::d etail;//define parameters vector<string> Img_names;bool Try_gpu = False;double Work_megapix = 0.6;//Image matches the resolution size, the area size of the image changes to work_megapix*100000double Seam_megapix = 0.1;//stitch pixel size Double Compose_megapix =0.6;//Stitching resolution FLOAT Conf_thresh = 1.f;//Two pictures from the confidence level of the same panorama wavecorrectkind wave_correct = detail::wave_correct_horiz;//waveform check, horizontal int expos_comp_type = exposurecompensator::gain_blocks;//Illumination compensation method, default is Gain_blocksfloat match_conf = 0.3f;//feature point detection confidence level, nearest neighbor match distance to the second nearest neighbor match distance, surf default is 0.65int Blend_type = blender::multi_band;//Fusion method, The default is multiband fusion float blend_strength = 5;//Fusion strength, 0-100. The default is 5.string result_name = "result.jpg";//output image filename int main () {clock_t   Start,finish;   Double totaltime; Start=clock (); int argc = 10;char* argv[] = {"1.jpg", "2.jpg", "3.jpg", "4.jpg", "5.jpg", "6.jpg", "7.jpg", "8.jpg", "9.jpg" "," 10.jpg "};for (int i = 0; i < argc; ++i) Img_names.push_back (argv[i]), int num_images = static_cast<int> (Img_names.size ());d ouble Work_scale = 1, Seam_scale = 1, Compose_scale = 1;//feature point detection and preprocessing (dimension scaling), then calculates the feature point of each graphic, and the feature point description sub cout<< "Finding Features ..." < <endl; ptr<featuresfinder> Finder;finder = new Surffeaturesfinder ()//////Use Surf feature points to detect Mat Full_img1,full_img, Img;vector <ImageFeatures> features (num_images);vector<mat> images (num_images);vector<size> full_img_sizes (num_images);d ouble seam_work_aspect = 1;for (int i = 0; i < num_images; ++i) {full_img1 = Imread (img_names[i]); Resize (fu ll_img1,full_img, size (400,300)); Full_img_sizes[i] = Full_img.size ();//calculation Work_scale, resize the image to an area of work_megapix*10^ 6 The following work_scale = Min (1.0, sqrt (Work_megapix * 1e6/full_img.size (). Area ())), Resize (full_img, IMG, size (), Work_scale, wor K_scale);//Resize the image to the area below work_megapix*10^6 Seam_scale = min (1.0, sqrt (Seam_megapix * 1e6/full_img.size (). Areas ())); Seam_work_aspect = seam_scale/work_scale;//Computes the feature point of the image, calculates the feature point descriptor, and sets the Img_idx to I (*finder) (IMG, features[i]); features[i ].img_idx = i;cout<< "Features in Image #" << i+1 << ":" << features[i].keypoints.size () <<en dl;//The source image resize to seam_megapix*10^6 and deposited in image[] Resize (full_img, IMG, Size (), Seam_scale, Seam_scale); Images[i] = Img.clone ();} Finder->collectgarbage (); Full_img.release (); Img.release ();//22 Matching of images cout<< "pairwise matching" << endl;//uses nearest neighbor and nearest neighbor match to match the feature points of any two images VECTOR&LT;MATCHESINFO&GT Pairwise_matches; Bestof2nearestmatcher Matcher (Try_gpu, match_conf);//nearest neighbor and sub-nearest neighbor Method Matcher (features, pairwise_matches); Match Matcher.collectgarbage () for every two images;//Combine all matches with confidence levels above the threshold into one set///leave only a picture that is determined to be from the same panorama vector<int> indices = Leavebiggestcomponent (features, pairwise_matches, Conf_thresh);vector<mat> img_subset;vector<string> Img_names_subset;vector<size> full_img_sizes_subset;for (size_t i = 0; i < indices.size (); ++i) {Img_names_ Subset.push_back (Img_names[indices[i]]); Img_subset.push_back (Images[indices[i]); full_img_sizes_subset.push_ Back (Full_img_sizes[indices[i]]);} Images = Img_subset;img_names = img_names_subset;full_img_sizes = full_img_sizes_subset;//Check whether the number of pictures still satisfies the requirements num_images = Static_cast<int> (Img_names.size ()); if (Num_images < 2) {cout<< "need more Images" <<endl;return-1 ;} Homographybasedestimator estimator;//based on the estimation of the vector<cameraparams> cameras;//camera Parameters Estimator (features, Pairwise_matches, cameras); for (size_t i = 0; i < CAMERAS.SIze (); ++i) {Mat r;cameras[i]. R.convertto (R, cv_32f); Cameras[i]. R = r;cout<< "Initial intrinsics #" << indices[i]+1 << ": \ n" << cameras[i]. K () <<endl;} Ptr<detail::bundleadjusterbase> adjuster;//Beam regulator parameter adjuster = new Detail::bundleadjusterray ();//Use bundle Adjustment (Light Bundle difference) method for all images of the camera parameter correction Adjuster->setconfthresh (CONF_THRESH);//Set configuration threshold mat_<uchar> Refine_mask = Mat::zeros (3, 3, cv_8u); Refine_mask (0,0) = 1;refine_mask (0,1) = 1;refine_mask (0,2) = 1;refine_mask (All) = 1;refine_mask ( ) = 1;adjuster->setrefinementmask (refine_mask);(*adjuster) (features, pairwise_matches, cameras);//Corrective/ The focal length is calculated to take the median and the focal length of all pictures and build the camera parameter, write the matrix to cameravector<double> focals;for (size_t i = 0; i < cameras.size (); ++i) {cout& lt;< "Camera #" << indices[i]+1 << ": \ n" << cameras[i]. K () <<endl;focals.push_back (cameras[i].focal);} Sort (Focals.begin (), Focals.end ()), Float warped_image_scale;if (focals.size ()% 2 = = 1) Warped_image_scale = static_cast <Float> (Focals[focals.size ()/2]) Elsewarped_image_scale = static_cast<float> (Focals[focals.size ()/2-1] + Focals[focals.size ()/2]) * 0.5f;///waveform correction vector<mat> rmats;for (size_t i = 0; i < cameras.size (); ++i) Rmats.push_ Back (Cameras[i]. R); Wavecorrect (Rmats, wave_correct);////waveform correction for (size_t i = 0; i < cameras.size (); ++i) Cameras[i]. R = rmats[i];cout<< "warping images ... "<<endl;vector<Point> Corners (num_images);//The Vertex vector<mat> masks_warped (num_images) after the unified coordinates; Vector<mat> images_warped (num_images);vector<size> sizes (num_images);vector<mat> Masks (num_ images);//Fusion mask//Prepare Image fusion mask for (int i = 0; i < num_images; ++i) {masks[i].create (Images[i].size (), cv_8u); Masks[i].setto (Scalar::all (255));} Curved image and Fusion mask ptr<warpercreator> warper_creator;warper_creator = new Cv::sphericalwarper (); Ptr<rotationwarper> warper = warper_creator->create (static_cast<float> (Warped_image_scale * SEAM_ Work_aspect)); for (int i = 0; I < Num_images; ++i) {mat_<float> K;cameras[i]. K (). ConvertTo (k, cv_32f); float swa = (float) seam_work_aspect; K (0,0) *= swa; K (0,2) *= swa; K (*=) SWA; K (*= swa;corners[i] = Warper->warp (Images[i], K, Cameras[i]. R, Inter_linear, Border_reflect, images_warped[i]);//calculate unified coordinate vertex sizes[i] = images_warped[i].size (); Warper->warp ( Masks[i], K, Cameras[i]. R, Inter_nearest, Border_constant, masks_warped[i]);//curved Current image}vector<mat> Images_warped_f (num_images); for (int i = 0; i < num_images; ++i) Images_warped[i].convertto (Images_warped_f[i], cv_32f); ptr<exposurecompensator> compensator = Exposurecompensator::createdefault (Expos_comp_type);//Set up a compensator for care compensation, The compensation method is gain_blockscompensator->feed (corners, images_warped, masks_warped);//Find seams ptr<seamfinder> seam_    Finder;seam_finder = new Detail::graphcutseamfinder (Graphcutseamfinderbase::cost_color); Seam_finder->find (Images_warped_f, corners, masks_warped);//Release unused memory Images.clear (); Images_warped.clear (); images_warped_f.clEar (); Masks.clear ();//////image fusion cout<< "compositing ..." <<endl; Mat img_warped, img_warped_s; Mat Dilated_mask, Seam_mask, Mask, mask_warped; ptr<blender> blender;double compose_work_aspect = 1;for (int img_idx = 0; img_idx < num_images; ++img_idx) {COUT&L t;< "Compositing image #" << indices[img_idx]+1<<endl;//because previously processed images are scaled in Work_scale, the image's internal// Corner (vertex after unified coordinates), mask (fused mask) needs to be recalculated//read the image and make necessary adjustments FULL_IMG1 = Imread (Img_names[img_idx]); Resize (full_img1,full_img , Size (400,300)); compose_scale = min (1.0, sqrt (Compose_megapix * 1e6/full_img.size (). Area ())); Compose_work_aspect = compose_scale/work_scale;//Update Curved Image scale Warped_image_scale *= static_cast<float> (compose_work_aspect); warper = Warper_creator->create (Warped_image_scale);//Update corners and sizesfor (int i = 0; i < num_images; ++i) {// Update camera following features cameras[i].focal *= compose_work_aspect;cameras[i].ppx *= compose_work_aspect;cameras[i].ppy *= compose_work _aspect;//update corners and sizessize sz = full_img_sizes[i]; if (Std::abs (compose_scale-1) > 1e-1) {sz.width = Cvround (Full_img_sizes[i].width * compose_scale); sz.height = CvRou nd (Full_img_sizes[i].height * compose_scale);} Mat K;cameras[i]. K (). ConvertTo (k, cv_32f); Rect ROI = Warper->warproi (sz, K, Cameras[i]. R); Corners[i] = roi.tl (); Sizes[i] = Roi.size ();} if (ABS (COMPOSE_SCALE-1) > 1e-1) Resize (full_img, IMG, Size (), Compose_scale, compose_scale); elseimg = Full_img;full_ Img.release (); Size img_size = Img.size (); Mat K;cameras[img_idx]. K (). ConvertTo (k, cv_32f);//Distort the current image Warper->warp (IMG, K, Cameras[img_idx]. R, Inter_linear, Border_reflect, img_warped);//distort Current image mask Mask.create (img_size, cv_8u); Mask.setto (Scalar::all (255)); Warper->warp (Mask, K, Cameras[img_idx]. R, Inter_nearest, Border_constant, mask_warped);//Exposure compensation compensator->apply (IMG_IDX, Corners[img_idx], img_warped, mask_warped); Img_warped.convertto (img_warped_s, cv_16s); Img_warped.release (); Img.release (); Mask.release (); Dilate (Masks_warped[img_idx], Dilated_mask, Mat ()); resize(Dilated_mask, Seam_mask, Mask_warped.size ()); mask_warped = seam_mask & mask_warped;//Initialize Blenderif (blender.empty ()) {blender = Blender::createdefault (Blend_type, TRY_GPU); Size DST_SZ = Resultroi (corners, sizes). Size (); float blend_width = sqrt (static_cast<float> (Dst_sz.area ())) * Blend_strength/100.f;if (Blend_width < 1.f) blender = Blender::createdefault (Blender::no, TRY_GPU); else { multibandblender* MB = dynamic_cast<multibandblender*> (static_cast<blender*> (Blender));mb-> Setnumbands (static_cast<int> (Ceil (blend_width)/log (2.))-1.); cout<< "Multi-Band blender, number of bands:" << mb->numbands () <<endl;} Determines the size of the final panorama based on the size of the corners vertex and image blender->prepare (corners, sizes);} Fused current Image blender->feed (img_warped_s, mask_warped, Corners[img_idx]);}  Mat result, Result_mask;blender->blend (result, result_mask); Imwrite (result_name, result);   Finish=clock ();   Totaltime= (Double) (Finish-start)/clocks_per_sec; cout<< "\" This program runs for "<<totaltime<< "Seconds! "<<endl;return 0;}

Any n image splicing _ effect very good _ computer Vision Big Job 1 Final edition