ISCE_INSAR/contrib/stack/stripmapStack/insarStack.py

#! /usr/bin/env python

#Author: Heresh Fattahi

import os
import sys
import h5py
import insarPair as insarPair
from numpy import median, float32, vstack

chunk_shape =(128,128)
dataType = float32

class insarStack:
    """    
        InsarStack object for a stack of InSAR data (multi-platform multi-track data).
        Method save2h5 creates a HDF5 file to store the stack data.
        Each platform-track is a group with three sub groups : observation, quality, geometry.
        Each sub-group may have different datasets. Some common datasets are:

        observations (3D) :  LOS, North, East, Up, RangeOffset, AzimuthOffset, ...
        quality  (3D)    :  coherence, uncertainty, ...
        geometry (2D or 3D)    :  incidence, heading angle, ...              
       
        All pairs for a given platform-track are required to have the same size.
        Pairs of different platforms-tracks may have different size.
        
 
    """  
    def __init__(self, name='insarStack', pairsDict = None):
        self.pairs = pairsDict

    def save2h5(self, output = 'data.h5', access_mode = 'w' , platform_tracks = None , ref_pixels = None , ref_pixel_method = 'average_coherence' ):
        '''
         h5OutName : Name of the HDF5 file for the InSAR stack

         platform_tracks : A list containing the platform_tracks to be stored 
                           in the HDF5 file. If None all platform_tracks are
                           exctracted from pairs. If pairs does not contain information
                           about the platform_tracks, then all pairs in the pairsDict are 
                           considered from a single platform single track.

         ref_pixel       : A dictionary containing refernce pixels for each platform_track.
                           eg: ref_pixDict = {'Sentinel1A/Track144':(500,300) , 'Sentinel1A/Track100':(440,700)}     
                               first pixel is in y direction (lines) and second pixel in x direction (columns)

        '''                    
        self.h5file = h5py.File(output, access_mode)
        self.platform_tracks = platform_tracks
        self.ref_pixels = ref_pixels
        if self.platform_tracks is None:
            self.get_platform_tracks()

        for platTrack in self.platform_tracks:
            print ('platform-track : ' , platTrack)
            group = self.h5file.create_group(platTrack)
            obsGroup = group.create_group('observations')
            qualityGroup = group.create_group('quality')
            geometryGroup = group.create_group('geometry')
            ################################
            # A class object for the platformTrack
            platTrackObj = platformTrack() 
            platTrackObj.getPairs(self.pairs, platTrack)
            platTrackObj.getSize()
            platTrackObj.getDatasetNames()
            ###############################
            # Get the reference pixel for a given platform/track
            if self.ref_pixels is not None:
               platTrackObj.ref_pixel = self.ref_pixels[platTrack]
            else:
               platTrackObj.ref_pixel = None
            ###############################
            # 3D datasets for observation quality (Coherence, uncertainty, ...)
            pairs = [pair for pair in platTrackObj.pairs.keys()]
            for dsName in platTrackObj.dsetQualityNames:
                print ('Create dataset for ', dsName)
                dsq = qualityGroup.create_dataset(dsName, shape=(platTrackObj.numPairs, platTrackObj.length, platTrackObj.width),
                        dtype=dataType) # , chunks=chunk_shape

                referenceTimes = [None]*platTrackObj.numPairs
                secondaryTimes = [None]*platTrackObj.numPairs
                for i in range(platTrackObj.numPairs):
                   data, metadata = platTrackObj.pairs[pairs[i]].read(dsName)
                   dsq[i,:,:] = data
                   reference , secondary = pairs[i]
                   referenceTimes[i] = reference.strftime('%Y-%m-%d %H:%M:%S').encode('utf8')
                   secondaryTimes[i] = secondary.strftime('%Y-%m-%d %H:%M:%S').encode('utf8')
            
            ###############################
            # store the pair times as a 2D dataset
            if len(platTrackObj.dsetQualityNames) > 0:
                piars_idx = vstack((referenceTimes,secondaryTimes)).T
                dsq = qualityGroup.create_dataset('pairs_idx', data=piars_idx, dtype=piars_idx.dtype)
            ###############################
            # if the reference pixel is not given let's choose a pixel with maximum average coherence
            #if platTrackObj.ref_pixel is None:
            #    platTrackObj.ref_pixel = self.choose_ref_pixel(platTrack , method == 'average_coherence')

            ###############################
            # 3D datasets for observations (possible datasets: unwrapped-phase, RangeOffset, AzimuthOffset, unwrapped-amplitude, etc)
            # There should be no limitation for storing any other possible observations.

            pairs = [pair for pair in platTrackObj.pairs.keys()]
             
            for dsName in platTrackObj.dsetObservationNames:
                print ('Create dataset for ', dsName)
                dso = obsGroup.create_dataset(dsName, shape=(platTrackObj.numPairs, platTrackObj.length, platTrackObj.width),
                        dtype=dataType) #, chunks=chunk_shape)
            
                referenceTimes = [None]*platTrackObj.numPairs
                secondaryTimes = [None]*platTrackObj.numPairs
                for i in range(platTrackObj.numPairs):
                   data, metadata = platTrackObj.pairs[pairs[i]].read(dsName)
                   #ds[i,:,:] = data - data[0, platTrackObj.ref_pixel[0] , platTrackObj.ref_pixel[1]]
                   dso[i,:,:] = data
                   reference , secondary = pairs[i]
                   referenceTimes[i] = reference.strftime('%Y-%m-%d %H:%M:%S').encode("ascii", "ignore")
                   secondaryTimes[i] = secondary.strftime('%Y-%m-%d %H:%M:%S').encode("ascii", "ignore")
            
            ###############################
            # A 2D dataset containing a 2D array of strings. First column 
            # is the reference time and second column the secondary time of pairs.
            if len(platTrackObj.dsetObservationNames) > 0:
                piars_idx = vstack((referenceTimes,secondaryTimes)).T
                dspairs = obsGroup.create_dataset('pairs_idx', data=piars_idx, dtype=piars_idx.dtype)
            ###################################
            for key,value in metadata.items():
                obsGroup.attrs[key] = value

            ###################################
            # 2D or 3D datasets for geometry (Lat, Lon, Heigt, Incidence, 
            # Heading, Bperp, ...). For a given platform from a specific 
            # track, a common viewing geometry is assumed. Therfore each 
            # of Lat, Lon, Height, Incidence and Heading can be stored as
            # 2D dataset. Baselines if provided should be 3D. 

            for dsName in platTrackObj.dsetGeometryNames:
                print ('Create dataset for ', dsName)
                pairs, length, width = platTrackObj.getSize_geometry(dsName)
                numPairs = len(pairs)
                dsg = geometryGroup.create_dataset(dsName, shape=(numPairs, length, width),
                        dtype=dataType) #, chunks=chunk_shape)
                
                for i in range(numPairs):
                   data, metadata = platTrackObj.pairs[pairs[i]].read(dsName)
                   dsg[i,:,:] = data

            for key,value in metadata.items():
                geometryGroup.attrs[key] = value

        
        self.h5file.close()

    def get_platform_tracks(self):

        self.platform_tracks = []
        for pair in self.pairs.keys():
            if self.pairs[pair].platform_track not in self.platform_tracks:
               self.platform_tracks.append(self.pairs[pair].platform_track)
        

    # def loadh5(self, platform_track , groupName='observation', datasetName='unwrapped', method = , method_par, )
         
    #     method     :  chunck       , block       , all
    #     method_par :  Chunck_size  , block_size  ,  

#    def choose_reference_pixel(self, platTrack , method):
        
        # compute average coherence of the 3D dataset
        # find the pixel with maximum value



#    def time_baseline_timeseries():

    

##################################
        
class platformTrack:

    def __init__(self, name='platformTrack'): #, pairDict = None):
        self.pairs = None
         
    def getPairs(self, pairDict, platTrack):
        pairs = pairDict.keys()
        self.pairs = {}
        for pair in pairs:
            if pairDict[pair].platform_track == platTrack:
               self.pairs[pair]=pairDict[pair]
    
    def getSize_geometry(self, dsName):
        pairs = self.pairs.keys()
        pairs2 = []
        width = []
        length = []
        files = []
        for pair in pairs:
            self.pairs[pair].get_metadata(dsName)
            if self.pairs[pair].length != 0 and self.pairs[pair].file not in files:
               files.append(self.pairs[pair].file)
               pairs2.append(pair)
               width.append(self.pairs[pair].width)
               length.append(self.pairs[pair].length)

        length = median(length)
        width  = median(width)
        return pairs2, length, width
 
    def getSize(self): 
        pairs = self.pairs.keys()
        self.numPairs = len(pairs)
        width = []
        length = []
        for pair in pairs:
            length.append(self.pairs[pair].length)
            width.append(self.pairs[pair].width)
        self.length = median(length)
        self.width  = median(width)

    def getDatasetNames(self): 
        # extract the name of the datasets which are actually the keys of 
        # observations, quality and geometry dictionaries.


        pairs = [pair for pair in self.pairs.keys()]
        # Assuming all pairs of a given platform-track have the same observations 
        # let's extract the keys of the observations of the first pair.
         
        if self.pairs[pairs[0]].observationsDict is not None: 
            self.dsetObservationNames = [k for k in self.pairs[pairs[0]].observationsDict.keys()]
        else:
            self.dsetObservationNames = []

        # Assuming all pairs of a given platform-track have the same quality files
        # let's extract the keys of the quality dictionary of the first pair. 
        if self.pairs[pairs[0]].qualityDict is not None:
            self.dsetQualityNames = [k for k in self.pairs[pairs[0]].qualityDict.keys()]                
        else:
            self.dsetQualityNames = []

        ##################
        # Despite the observation and quality files, the geometry may not exist
        # for all pairs. Therfore we need to look at all pairs and get possible 
        # dataset names.
        self.dsetGeometryNames = []
        for pair in pairs:
            if self.pairs[pair].geometryDict  is not None:
               keys = [k for k in self.pairs[pair].geometryDict.keys()]       
               self.dsetGeometryNames = list(set(self.dsetGeometryNames) | set(keys))