# -*- coding: utf-8 -*- #==========================================================# # Extract shorelines from Landsat images #==========================================================# # Initial settings import os import numpy as np import matplotlib.pyplot as plt import ee import pdb # other modules from osgeo import gdal, ogr, osr import pickle import matplotlib.cm as cm from pylab import ginput # image processing modules import skimage.filters as filters import skimage.exposure as exposure import skimage.transform as transform import sklearn.decomposition as decomposition import skimage.measure as measure # import own modules import functions.utils as utils import functions.sds as sds # some settings np.seterr(all='ignore') # raise/ignore divisions by 0 and nans plt.rcParams['axes.grid'] = True plt.rcParams['figure.max_open_warning'] = 100 ee.Initialize() # parameters cloud_thresh = 0.5 # threshold for cloud cover plot_bool = False # if you want the plots prob_high = 99.9 # upper probability to clip and rescale pixel intensity min_contour_points = 100# minimum number of points contained in each water line output_epsg = 28356 # GDA94 / MGA Zone 56 # load metadata (timestamps and epsg code) for the collection satname = 'L8' #sitename = 'NARRA' sitename = 'OLDBAR' filepath = os.path.join(os.getcwd(), 'data', satname, sitename) with open(os.path.join(filepath, sitename + '_timestamps' + '.pkl'), 'rb') as f: timestamps = pickle.load(f) timestamps_sorted = sorted(timestamps) # sort timestamps since images are sorted in directory with open(os.path.join(filepath, sitename + '_epsgcode' + '.pkl'), 'rb') as f: input_epsg = pickle.load(f) with open(os.path.join(filepath, sitename + '_refpoints' + '.pkl'), 'rb') as f: refpoints = pickle.load(f) # path to images file_path_pan = os.path.join(os.getcwd(), 'data', satname, sitename, 'pan') file_path_ms = os.path.join(os.getcwd(), 'data', satname, sitename, 'ms') file_names_pan = os.listdir(file_path_pan) file_names_ms = os.listdir(file_path_ms) N = len(file_names_pan) # initialise some variables cloud_cover_ts = [] date_acquired_ts = [] idx_skipped = [] idx_nocloud = [] t = [] shorelines = [] for i in range(N): # read pan image fn_pan = os.path.join(file_path_pan, file_names_pan[i]) data = gdal.Open(fn_pan, gdal.GA_ReadOnly) georef = np.array(data.GetGeoTransform()) bands = [data.GetRasterBand(i + 1).ReadAsArray() for i in range(data.RasterCount)] im_pan = np.stack(bands, 2)[:,:,0] # read ms image fn_ms = os.path.join(file_path_ms, file_names_ms[i]) data = gdal.Open(fn_ms, gdal.GA_ReadOnly) bands = [data.GetRasterBand(i + 1).ReadAsArray() for i in range(data.RasterCount)] im_ms = np.stack(bands, 2) # cloud mask im_qa = im_ms[:,:,5] cloud_mask = sds.create_cloud_mask(im_qa, satname, plot_bool) cloud_mask = transform.resize(cloud_mask, (im_pan.shape[0], im_pan.shape[1]), order=0, preserve_range=True, mode='constant').astype('bool_') # resize the image using bilinear interpolation (order 1) im_ms = transform.resize(im_ms,(im_pan.shape[0], im_pan.shape[1]), order=1, preserve_range=True, mode='constant') # check if -inf or nan values and add to cloud mask im_inf = np.isin(im_ms[:,:,0], -np.inf) im_nan = np.isnan(im_ms[:,:,0]) cloud_mask = np.logical_or(np.logical_or(cloud_mask, im_inf), im_nan) cloud_cover = sum(sum(cloud_mask.astype(int)))/(cloud_mask.shape[0]*cloud_mask.shape[1]) if cloud_cover > cloud_thresh: print('skipped cloud ' + str(i)) idx_skipped.append(i) continue idx_nocloud.append(i) # check if image for that date is already present and keep the one with less clouds if file_names_pan[i][len(satname)+1+len(sitename)+1:len(satname)+1+len(sitename)+1+10] in date_acquired_ts: idx_samedate = utils.find_indices(date_acquired_ts, lambda e : e == file_names_pan[i][9:19]) idx_samedate = idx_samedate[0] print(str(cloud_cover) + ' - ' + str(cloud_cover_ts[idx_samedate])) if cloud_cover >= cloud_cover_ts[idx_samedate]: print('skipped double ' + str(i)) idx_skipped.append(i) continue else: del shorelines[idx_samedate] del t[idx_samedate] del cloud_cover_ts[idx_samedate] del date_acquired_ts[idx_samedate] print('deleted ' + str(idx_samedate)) # rescale intensities im_ms = sds.rescale_image_intensity(im_ms, cloud_mask, prob_high, plot_bool) im_pan = sds.rescale_image_intensity(im_pan, cloud_mask, prob_high, plot_bool) # pansharpen rgb image im_ms_ps = sds.pansharpen(im_ms[:,:,[0,1,2]], im_pan, cloud_mask, plot_bool) # add down-sized bands for NIR and SWIR (since pansharpening is not possible) im_ms_ps = np.append(im_ms_ps, im_ms[:,:,[3,4]], axis=2) # calculate NDWI im_ndwi = sds.nd_index(im_ms_ps[:,:,3], im_ms_ps[:,:,1], cloud_mask, plot_bool) # detect edges wl_pix = sds.find_wl_contours(im_ndwi, cloud_mask, min_contour_points, plot_bool) # convert from pixels to world coordinates wl_coords = sds.convert_pix2world(wl_pix, georef) # convert to output epsg spatial reference wl = sds.convert_epsg(wl_coords, input_epsg, output_epsg) # plot a figure to select the correct water line and discard cloudy images plt.figure() cmap = cm.get_cmap('jet') plt.subplot(121) plt.imshow(im_ms_ps[:,:,[2,1,0]]) for j,contour in enumerate(wl_pix): colours = cmap(np.linspace(0, 1, num=len(wl_pix))) plt.plot(contour[:, 1], contour[:, 0], linewidth=2, color=colours[j,:]) plt.axis('image') plt.title(file_names_pan[i]) plt.subplot(122) centroids = [] for j,contour in enumerate(wl): colours = cmap(np.linspace(0, 1, num=len(wl))) centroids.append([np.mean(contour[:, 0]),np.mean(contour[:, 1])]) plt.plot(contour[:, 0], contour[:, 1], linewidth=2, color=colours[j,:]) plt.plot(np.mean(contour[:, 0]), np.mean(contour[:, 1]), 'o', color=colours[j,:]) plt.plot(refpoints[:,0], refpoints[:,1], 'k.') plt.axis('equal') plt.title(file_names_pan[i]) mng = plt.get_current_fig_manager() mng.window.showMaximized() plt.tight_layout() plt.draw() # click on the left image to discard, otherwise on the closest centroid in the right image pt_in = np.array(ginput(n=1, timeout=1000)) if pt_in[0][0] < 10000: print('skipped manual ' + str(i)) idx_skipped.append(i) continue # get contour that was selected (clock closest to centroid) dist_centroid = [np.linalg.norm(_ - pt_in) for _ in centroids] shorelines.append(wl[np.argmin(dist_centroid)]) t.append(timestamps_sorted[i]) cloud_cover_ts.append(cloud_cover) date_acquired_ts.append(file_names_pan[i][9:19]) #plt.figure() #plt.axis('equal') #for j in range(len(shorelines)): # plt.plot(shorelines[j][:,0], shorelines[j][:,1]) #plt.draw() output = {'t':t, 'shorelines':shorelines, 'cloud_cover':cloud_cover_ts} with open(os.path.join(filepath, sitename + '_output' + '.pkl'), 'wb') as f: pickle.dump(output, f) with open(os.path.join(filepath, sitename + '_skipped' + '.pkl'), 'wb') as f: pickle.dump(idx_skipped, f) with open(os.path.join(filepath, sitename + '_idxnocloud' + '.pkl'), 'wb') as f: pickle.dump(idx_nocloud, f)