How to use the histomicstk.segmentation.label function in histomicstk

# run adaptive multi-scale LoG filter
    im_log_max, im_sigma_max = htk_shape_filters.cdog(
        im_nuclei_stain, im_nuclei_fgnd_mask,
        sigma_min=min_radius / np.sqrt(2),
        sigma_max=max_radius / np.sqrt(2)
    )

    # apply local maximum clustering
    im_nuclei_seg_mask, seeds, maxima = htk.segmentation.nuclear.max_clustering(
        im_log_max, im_nuclei_fgnd_mask, local_max_search_radius)

    if seeds is None:
        return im_nuclei_seg_mask

    # split any objects with disconnected fragments
    im_nuclei_seg_mask = htk.segmentation.label.split(im_nuclei_seg_mask,
                                                      conn=8)

    # filter out small objects
    im_nuclei_seg_mask = htk.segmentation.label.area_open(
        im_nuclei_seg_mask, min_nucleus_area).astype(np.int)

    return im_nuclei_seg_mask

sigma_max=max_radius / np.sqrt(2)
    )

    # apply local maximum clustering
    im_nuclei_seg_mask, seeds, maxima = htk.segmentation.nuclear.max_clustering(
        im_log_max, im_nuclei_fgnd_mask, local_max_search_radius)

    if seeds is None:
        return im_nuclei_seg_mask

    # split any objects with disconnected fragments
    im_nuclei_seg_mask = htk.segmentation.label.split(im_nuclei_seg_mask,
                                                      conn=8)

    # filter out small objects
    im_nuclei_seg_mask = htk.segmentation.label.area_open(
        im_nuclei_seg_mask, min_nucleus_area).astype(np.int)

    return im_nuclei_seg_mask

X, Y, Min, Max = seed_contours(I, Delta)

    # trace contours from seeds
    cXs, cYs = trace_contours(I, X, Y, Min, Max, MaxLength=255)

    # score successfully traced contours
    Scores = score_contours(I, cXs, cYs)

    # construct label image from scored contours
    Label = label_contour(I.shape, cXs, cYs, Scores)

    # compact contours to remove spurs - the paper calls this "optimization"
    Label = label.compact(Label, Compaction)

    # cleanup label image
    Label = label.split(Label)
    Label = label.area_open(Label, MinArea)
    Label = label.width_open(Label, MinWidth)

    # split objects with concavities
    Label = split_concavities(Label, MinDepth, MinConcavity)

    return Label

"""

    # identify contour seed points
    X, Y, Min, Max = seed_contours(I, Delta)

    # trace contours from seeds
    cXs, cYs = trace_contours(I, X, Y, Min, Max, MaxLength=255)

    # score successfully traced contours
    Scores = score_contours(I, cXs, cYs)

    # construct label image from scored contours
    Label = label_contour(I.shape, cXs, cYs, Scores)

    # compact contours to remove spurs - the paper calls this "optimization"
    Label = label.compact(Label, Compaction)

    # cleanup label image
    Label = label.split(Label)
    Label = label.area_open(Label, MinArea)
    Label = label.width_open(Label, MinWidth)

    # split objects with concavities
    Label = split_concavities(Label, MinDepth, MinConcavity)

    return Label

def create_tile_nuclei_boundary_annotations(im_nuclei_seg_mask, tile_info):

    nuclei_annot_list = []

    gx = tile_info['gx']
    gy = tile_info['gy']
    wfrac = tile_info['gwidth'] / np.double(tile_info['width'])
    hfrac = tile_info['gheight'] / np.double(tile_info['height'])

    by, bx = htk_seg.label.trace_object_boundaries(im_nuclei_seg_mask,
                                                   trace_all=True)

    for i in range(len(bx)):

        # get boundary points and convert to base pixel space
        num_points = len(bx[i])

        if num_points < 3:
            continue

        cur_points = np.zeros((num_points, 3))
        cur_points[:, 0] = np.round(gx + bx[i] * wfrac, 2)
        cur_points[:, 1] = np.round(gy + by[i] * hfrac, 2)
        cur_points = cur_points.tolist()

        # create annotation json

if region_props[i].label in valid_superpixel:
            min_row, max_row, min_col, max_col = \
                get_boundary_bounds(region_props[i].bbox, 0, n_rows, n_cols)

            # grab label mask
            lmask = (
                im_label[min_row:max_row, min_col:max_col] ==
                region_props[i].label
            ).astype(np.bool)

            mask = np.zeros(
                (lmask.shape[0] + 2, lmask.shape[1] + 2), dtype=np.bool)
            mask[1:-1, 1:-1] = lmask

            # find boundaries
            bx, by = htk_seg.label.trace_object_boundaries(mask)
            bx = bx[0] + min_row
            by = by[0] + min_col

            with np.errstate(invalid='ignore'):
                # remove redundant points
                mby, mbx = htk_utils.merge_colinear(
                    by.astype(float), bx.astype(float))

            # get superpixel boundary at highest-res
            y_brs.append(mbx + top)
            x_brs.append(mby + left)
            cen_x, cen_y = region_props[i].centroid

            # get superpixel centers at highest-res
            y_cent.append(
                round((cen_x + top), 1))

max(0, X[i]-np.ceil(MaxLength/2.0)):
              min(I.shape[1]+1, X[i]+np.ceil(MaxLength/2.0))+1]

        # binary threshold corresponding to seed pixel 'i'
        W = (W <= Max[i]) & (W >= Min[i])

        # embed with center pixel in middle of padded window
        Embed = np.zeros((W.shape[0]+2, W.shape[1]+2), dtype=np.bool)
        Embed[1:-1, 1:-1] = W

        # calculate location of (X[i], Y[i]) in 'Embed'
        pX = X[i] - max(0, X[i]-np.ceil(MaxLength/2.0)) + 1
        pY = Y[i] - max(0, Y[i]-np.ceil(MaxLength/2.0)) + 1

        # trace boundary, check stopping condition, append to list of contours
        cX, cY = label.trace_object_boundaries(Embed, conn=4,
                                               x_start=pX, y_start=pY,
                                               MaxLength=MaxLength)
        if cX[0][0] == cX[0][-1] and cY[0][0] == cY[0][-1] and\
                len(cX[0]) <= MaxLength:

            # add window offset to contour coordinates
            cX[0] = [
                x + max(0, X[i]-np.ceil(MaxLength/2.0)) - 1 for x in cX[0]
            ]
            cY[0] = [
                y + max(0, Y[i]-np.ceil(MaxLength/2.0)) - 1 for y in cY[0]
            ]

            # append to list of candidate contours
            cXs.append(np.array(cX[0], dtype=np.uint32))
            cYs.append(np.array(cY[0], dtype=np.uint32))

def create_tile_nuclei_boundary_annotations(im_nuclei_seg_mask, tile_info):

    nuclei_annot_list = []

    gx = tile_info['gx']
    gy = tile_info['gy']
    wfrac = tile_info['gwidth'] / np.double(tile_info['width'])
    hfrac = tile_info['gheight'] / np.double(tile_info['height'])

    by, bx = htk_seg.label.trace_object_boundaries(im_nuclei_seg_mask,
                                                   trace_all=True)

    for i in range(len(bx)):

        # get boundary points and convert to base pixel space
        num_points = len(bx[i])

        if num_points < 3:
            continue

        cur_points = np.zeros((num_points, 3))
        cur_points[:, 0] = np.round(gx + bx[i] * wfrac, 2)
        cur_points[:, 1] = np.round(gy + by[i] * hfrac, 2)
        cur_points = cur_points.tolist()

        # create annotation json

def create_tile_nuclei_boundary_annotations(im_nuclei_seg_mask, tile_info):

    nuclei_annot_list = []

    gx = tile_info['gx']
    gy = tile_info['gy']
    wfrac = tile_info['gwidth'] / np.double(tile_info['width'])
    hfrac = tile_info['gheight'] / np.double(tile_info['height'])

    by, bx = htk_seg.label.trace_object_boundaries(im_nuclei_seg_mask,
                                                   trace_all=True)

    for i in range(len(bx)):

        # get boundary points and convert to base pixel space
        num_points = len(bx[i])

        if num_points < 3:
            continue

        cur_points = np.zeros((num_points, 3))
        cur_points[:, 0] = np.round(gx + bx[i] * wfrac, 2)
        cur_points[:, 1] = np.round(gy + by[i] * hfrac, 2)
        cur_points = cur_points.tolist()

        # create annotation json

im_nuclei_fgnd_mask = im_nuclei_stain < args.foreground_threshold

    # segment nuclei
    im_nuclei_seg_mask = htk_nuclear.detect_nuclei_kofahi(
        im_nuclei_stain,
        im_nuclei_fgnd_mask,
        args.min_radius,
        args.max_radius,
        args.min_nucleus_area,
        args.local_max_search_radius
    )

    # Delete border nuclei
    if args.ignore_border_nuclei is True:

        im_nuclei_seg_mask = htk_seg_label.delete_border(im_nuclei_seg_mask)

    # generate nuclei annotations
    nuclei_annot_list = []

    flag_nuclei_found = np.any(im_nuclei_seg_mask)

    if flag_nuclei_found:
        nuclei_annot_list = cli_utils.create_tile_nuclei_annotations(
            im_nuclei_seg_mask, tile_info, args.nuclei_annotation_format)

    return nuclei_annot_list