Exemplo n.º 1
0
def build_map(proj, group_list, group_index):
    # lookup ned reference
    ref_node = getNode("/config/ned_reference", True)
    ref = [
        ref_node.getFloat('lat_deg'),
        ref_node.getFloat('lon_deg'),
        ref_node.getFloat('alt_m')
    ]

    log("Loading optimized match points ...")
    matches = pickle.load(
        open(os.path.join(proj.analysis_dir, "matches_grouped"), "rb"))

    # initialize temporary structures for vanity stats
    for image in proj.image_list:
        image.sum_values = 0.0
        image.sum_count = 0.0
        image.max_z = -9999.0
        image.min_z = 9999.0

    # elevation stats
    log("Computing stats...")
    ned_list = []
    for match in matches:
        if match[1] == group_index:  # used by current group
            ned_list.append(match[0])
    avg = -np.mean(np.array(ned_list)[:, 2])
    std = np.std(np.array(ned_list)[:, 2])
    log("Average elevation: %.2f" % avg)
    log("Standard deviation: %.2f" % std)

    # sort through points
    log('Reading feature locations from optimized match points ...')
    raw_points = []
    raw_values = []
    for match in matches:
        if match[1] == group_index:  # used by current group
            ned = match[0]
            diff = abs(-ned[2] - avg)
            if diff < 10 * std:
                raw_points.append([ned[1], ned[0]])
                raw_values.append(ned[2])
                for m in match[2:]:
                    if proj.image_list[m[0]].name in group_list[group_index]:
                        image = proj.image_list[m[0]]
                        z = -ned[2]
                        image.sum_values += z
                        image.sum_count += 1
                        if z < image.min_z:
                            image.min_z = z
                            #print(min_z, match)
                        if z > image.max_z:
                            image.max_z = z
                            #print(max_z, match)
            else:
                log("Discarding match with excessive altitude:", match)

    # save the surface definition as a separate file
    models_dir = os.path.join(proj.analysis_dir, 'models')
    if not os.path.exists(models_dir):
        log("Notice: creating models directory =", models_dir)
        os.makedirs(models_dir)
    surface = {'points': raw_points, 'values': raw_values}
    pickle.dump(
        surface,
        open(os.path.join(proj.analysis_dir, 'models', 'surface.bin'), "wb"))

    log('Generating Delaunay mesh and interpolator ...')
    global_tri_list = scipy.spatial.Delaunay(np.array(raw_points))
    interp = scipy.interpolate.LinearNDInterpolator(global_tri_list,
                                                    raw_values)

    for image in proj.image_list:
        if image.sum_count > 0:
            image.z_avg = image.sum_values / float(image.sum_count)
            # log(image.name, 'avg elev:', image.z_avg)
        else:
            image.z_avg = 0

    # compute the uv grid for each image and project each point out into
    # ned space, then intersect each vector with the srtm / ground /
    # delauney surface.

    #for group in group_list:
    if True:
        group = group_list[group_index]
        #if len(group) < 3:
        #    continue
        for name in group:
            image = proj.findImageByName(name)
            log(image.name, image.z_avg)
            width, height = camera.get_image_params()
            # scale the K matrix if we have scaled the images
            K = camera.get_K(optimized=True)
            IK = np.linalg.inv(K)

            grid_list = []
            u_list = np.linspace(0, width, grid_steps + 1)
            v_list = np.linspace(0, height, grid_steps + 1)
            #print "u_list:", u_list
            #print "v_list:", v_list
            for v in v_list:
                for u in u_list:
                    grid_list.append([u, v])
            #print 'grid_list:', grid_list
            image.distorted_uv = proj.redistort(grid_list, optimized=True)

            if use_direct_pose:
                proj_list = project.projectVectors(IK, image.get_body2ned(),
                                                   image.get_cam2body(),
                                                   grid_list)
            else:
                #print(image.get_body2ned(opt=True))
                proj_list = project.projectVectors(
                    IK, image.get_body2ned(opt=True), image.get_cam2body(),
                    grid_list)
            #print 'proj_list:', proj_list

            if use_direct_pose:
                ned, ypr, quat = image.get_camera_pose()
            else:
                ned, ypr, quat = image.get_camera_pose(opt=True)
            #print('cam orig:', image.camera_pose['ned'], 'optimized:', ned)
            if force_ground_elevation_m:
                pts_ned = project.intersectVectorsWithGroundPlane(
                    ned, force_ground_elevation_m, proj_list)
            elif use_srtm_surface:
                # setup SRTM ground interpolator
                from lib import srtm
                srtm.initialize(ref, 6000, 6000, 30)
                pts_ned = srtm.interpolate_vectors(ned, proj_list)
            elif False:
                # this never seemed that productive
                print(image.name, image.z_avg)
                pts_ned = project.intersectVectorsWithGroundPlane(
                    ned, image.z_avg, proj_list)
            elif True:
                # intersect with our polygon surface approximation
                pts_ned = intersect_vectors(interp, ned, proj_list,
                                            -image.z_avg)
            elif False:
                # (moving away from the binned surface approach in this
                # script towards the above delauney interpolation
                # approach)
                # intersect with 2d binned surface approximation
                pts_ned = bin2d.intersect_vectors(interp, ned, proj_list,
                                                  -image.z_avg)

            #print(image.name, "pts_3d (ned):\n", pts_ned)

            # convert ned to xyz and stash the result for each image
            image.grid_list = []
            for p in pts_ned:
                image.grid_list.append([p[1], p[0], -p[2]])

    # generate the panda3d egg models
    dir_node = getNode('/config/directories', True)
    img_src_dir = dir_node.getString('images_source')
    panda3d.generate_from_grid(proj,
                               group_list[group_index],
                               src_dir=img_src_dir,
                               analysis_dir=proj.analysis_dir,
                               resolution=texture_resolution)
Exemplo n.º 2
0
                                                image.get_body2ned(opt=True),
                                                image.get_cam2body(),
                                                grid_list )
        #print 'proj_list:', proj_list

        if args.direct:
            ned, ypr, quat = image.get_camera_pose()
        else:
            ned, ypr, quat = image.get_camera_pose(opt=True)
        #print('cam orig:', image.camera_pose['ned'], 'optimized:', ned)
        if args.ground:
            pts_ned = project.intersectVectorsWithGroundPlane(ned,
                                                              args.ground,
                                                              proj_list)
        elif args.srtm:
            pts_ned = srtm.interpolate_vectors(ned, proj_list)
        else:
            # intersect with our polygon surface approximation
            pts_ned = intersect_vectors(ned, proj_list, -image.z_avg)
            
        #print(image.name, "pts_3d (ned):\n", pts_ned)

        # convert ned to xyz and stash the result for each image
        image.grid_list = []
        for p in pts_ned:
            image.fit_xy.append([p[1], p[0]])
            image.fit_z.append(-p[2])
            image.fit_edge.append(True)
        image.fit_uv = distorted_uv
        print('len:', len(image.fit_xy), len(image.fit_z), len(image.fit_uv))
dist_coeffs = proj.cam.get_dist_coeffs()
for image in proj.image_list:
    print image.name
    scale = float(image.width) / float(camw)
    K = proj.cam.get_K(scale)
    IK = np.linalg.inv(K)
    corner_list = []
    corner_list.append([0, 0])
    corner_list.append([image.width, 0])
    corner_list.append([0, image.height])
    corner_list.append([image.width, image.height])

    proj_list = project.projectVectors(IK, image, corner_list, pose=args.pose)
    #print "proj_list:\n", proj_list
    if args.pose == 'direct':
        pts_ned = srtm.interpolate_vectors(image.camera_pose, proj_list)
    elif args.pose == 'sba':
        pts_ned = srtm.interpolate_vectors(image.camera_pose_sba, proj_list)
    # print "pts (ned):\n", pts_ned

    image.corner_list_ned = []
    image.corner_list_lla = []
    image.corner_list_xy = []
    for ned in pts_ned:
        #print p
        image.corner_list_ned.append([ned[0], ned[1]])
        image.corner_list_lla.append(
            navpy.ned2lla([ned], ref[0], ref[1], ref[2]))
        image.corner_list_xy.append([ned[1], ned[0]])

dst_dir = proj.project_dir + "/Warped/"