[14] | 1 | #!/bin/env python |
---|
| 2 | # coding: utf-8 |
---|
| 3 | # Un programme pour exploiter les données des mires |
---|
| 4 | # -- Frédéric Meynadier |
---|
| 5 | |
---|
| 6 | import string |
---|
| 7 | from numpy import * |
---|
| 8 | |
---|
| 9 | from functions import * |
---|
| 10 | |
---|
| 11 | from optparse import OptionParser |
---|
| 12 | parser = OptionParser() |
---|
| 13 | parser.add_option("-c","--catalog",dest="catalogfile", |
---|
| 14 | help="Name of the catalog file") |
---|
| 15 | parser.add_option("-o","--outputfits",dest="outputfits", |
---|
| 16 | default="output.fits", |
---|
| 17 | help="Name of the output fits file") |
---|
| 18 | |
---|
| 19 | |
---|
| 20 | (options, args) = parser.parse_args () |
---|
| 21 | |
---|
| 22 | # Donner les coordonnées approximatives des 4 repÚres |
---|
| 23 | |
---|
| 24 | rep = zeros((4,2)) |
---|
| 25 | i = 0 |
---|
| 26 | for line in open('reperes.txt'): |
---|
| 27 | if ((line[0] != "#") and (line != '\n')): |
---|
| 28 | spl = string.split(line) |
---|
| 29 | rep[i][0] = float(spl[0]) |
---|
| 30 | rep[i][1] = float(spl[1]) |
---|
| 31 | i+=1 |
---|
| 32 | |
---|
| 33 | # Lecture du fichier catalogue |
---|
| 34 | fields = ["FLUX_MAX", |
---|
| 35 | "X_IMAGE", |
---|
| 36 | "Y_IMAGE", |
---|
| 37 | "THETA_IMAGE", |
---|
| 38 | "ELONGATION", |
---|
| 39 | "ELLIPTICITY", |
---|
| 40 | "FWHM_IMAGE", |
---|
| 41 | "ERRX2_IMAGE", |
---|
| 42 | "ERRY2_IMAGE"] |
---|
| 43 | |
---|
| 44 | catalog = [] |
---|
| 45 | for line in open(options.catalogfile): |
---|
| 46 | if ((line[0] != "#") and (line != '\n')): |
---|
| 47 | tmp = {} |
---|
| 48 | spl = string.split(line) |
---|
| 49 | for i in range(len(fields)): |
---|
| 50 | tmp[fields[i]] = float(spl[i]) |
---|
| 51 | catalog.append(tmp) |
---|
| 52 | |
---|
| 53 | |
---|
| 54 | # Recherche de la position des repÚres selon catalogue - attention aux doublons |
---|
| 55 | real_rep = zeros((4,2)) |
---|
| 56 | for i in range(4): |
---|
| 57 | detections = 0 |
---|
| 58 | for source in catalog: |
---|
| 59 | pos_source = array([source['X_IMAGE'],source['Y_IMAGE']]) |
---|
| 60 | if ((modulus(rep[i,:] - pos_source)) < 5): |
---|
| 61 | detections +=1 |
---|
| 62 | real_rep[i,:] = pos_source |
---|
| 63 | if (detections !=1): |
---|
| 64 | print "Attention ! ambiguïté pour repÚre ",i,\ |
---|
| 65 | ", nb de détections :",detections |
---|
| 66 | |
---|
| 67 | |
---|
| 68 | # Détermination du centre : barycentre des 4 repÚres |
---|
| 69 | center = array([sum(real_rep[:,0])/4,sum(real_rep[:,1])/4]) |
---|
| 70 | |
---|
| 71 | # Vecteurs unitaires de référence |
---|
| 72 | vec_i = ((real_rep[0,:] - real_rep[1,:])/20 \ |
---|
| 73 | + (real_rep[3,:] - real_rep[2,:])/20) / 2 |
---|
| 74 | vec_j = ((real_rep[1,:] - real_rep[2,:])/20 \ |
---|
| 75 | + (real_rep[0,:] - real_rep[3,:])/20) / 2 |
---|
| 76 | |
---|
| 77 | # Détection des noeuds de grille les plus proches de chaque source |
---|
| 78 | vec_i_2 = dot(vec_i,vec_i) |
---|
| 79 | vec_j_2 = dot(vec_j,vec_j) |
---|
| 80 | for source in catalog: |
---|
| 81 | pos_source = array([source['X_IMAGE'],source['Y_IMAGE']])-center |
---|
| 82 | # dans le référentiel grille (en projetant sur les vecteurs unitaires) : |
---|
| 83 | pos_source_grid = array([dot(pos_source,vec_i)/vec_i_2, |
---|
| 84 | dot(pos_source,vec_j)/vec_j_2]) |
---|
| 85 | # on arrondit au noeud le plus proche |
---|
| 86 | knot = pos_source_grid.round() |
---|
| 87 | # et on calcule la distance entre les deux |
---|
| 88 | r = modulus(pos_source_grid - knot) |
---|
| 89 | # enregistrement des résultats |
---|
| 90 | source['I_GRID'] = knot[0] |
---|
| 91 | source['J_GRID'] = knot[1] |
---|
| 92 | source['KNOT_DIST'] = r |
---|
| 93 | #if (modulus(knot) <1): |
---|
| 94 | # print source['I_GRID'], source['J_GRID'], source['KNOT_DIST'], pos_source_grid, pos_source,source['X_IMAGE'], source['Y_IMAGE'] |
---|
| 95 | |
---|
| 96 | # Chasse aux doublons |
---|
| 97 | # tableau 81*81 contenant : |
---|
| 98 | # champ 0 : nb de sources rattachée au noeud |
---|
| 99 | # champ 1 : distance actuelle entre le noeud et la source |
---|
| 100 | # champ 2 : index courant de la source dans catalog |
---|
| 101 | # champ 3 : distance moyenne aux voisins (calculé plus tard) |
---|
| 102 | lookup_sources = zeros((81,81,4)) |
---|
| 103 | for si in range(len(catalog)): |
---|
| 104 | source = catalog[si] |
---|
| 105 | i = source['I_GRID']+40 |
---|
| 106 | j = source['J_GRID']+40 |
---|
| 107 | if ((lookup_sources[i,j,0] < 1) |
---|
| 108 | or (lookup_sources[i,j,1] > source['KNOT_DIST'])): |
---|
| 109 | lookup_sources[i,j,2] = si |
---|
| 110 | lookup_sources[i,j,1] = source['KNOT_DIST'] |
---|
| 111 | lookup_sources[i,j,0] += 1 |
---|
| 112 | |
---|
| 113 | # On élimine le noeud (1,1), trop prÚs du "P" |
---|
| 114 | lookup_sources[41,41,0] = 0 |
---|
| 115 | |
---|
| 116 | # Calcul de la distance moyenne entre noeuds voisins |
---|
| 117 | |
---|
| 118 | for i in range(1,80): |
---|
| 119 | for j in range(1,80): |
---|
| 120 | if (lookup_sources[i,j,0] > 0): |
---|
| 121 | knot_idx = int(lookup_sources[i,j,2]) |
---|
| 122 | catalog[knot_idx]['MEAN_DIST_TO_NGBRS'] = 0 |
---|
| 123 | nb_voisin = 0 |
---|
| 124 | distance = 0 |
---|
| 125 | pos_source = array([catalog[knot_idx]['X_IMAGE'], |
---|
| 126 | catalog[knot_idx]['Y_IMAGE']]) |
---|
| 127 | for shifts in [[0,1],[1,0],[-1,0],[0,-1]]: |
---|
| 128 | i_voisin = i + shifts[0] |
---|
| 129 | j_voisin = j + shifts[1] |
---|
| 130 | if (lookup_sources[i_voisin,j_voisin,0] > 0): |
---|
| 131 | neigh_idx = int(lookup_sources[i_voisin,j_voisin,2]) |
---|
| 132 | nb_voisin +=1 |
---|
| 133 | pos_voisin = array([catalog[neigh_idx]['X_IMAGE'], |
---|
| 134 | catalog[neigh_idx]['Y_IMAGE']]) |
---|
| 135 | distance += modulus(pos_source-pos_voisin) |
---|
| 136 | if (nb_voisin !=0): |
---|
| 137 | catalog[knot_idx]['MEAN_DIST_TO_NGBRS'] = distance / nb_voisin |
---|
| 138 | lookup_sources[i,j,3] = distance / nb_voisin |
---|
| 139 | |
---|
| 140 | cleaned_catalog = [] |
---|
| 141 | for i in range(81): |
---|
| 142 | for j in range(81): |
---|
| 143 | if (lookup_sources[i,j,0] > 0): |
---|
| 144 | cleaned_catalog.append(catalog[int(lookup_sources[i,j,2])]) |
---|
| 145 | |
---|
| 146 | # Sortie du catalogue nettoyé des doublons |
---|
| 147 | |
---|
| 148 | of = open('clean.cat','w') |
---|
| 149 | |
---|
| 150 | fields = ["FLUX_MAX", |
---|
| 151 | "X_IMAGE", |
---|
| 152 | "Y_IMAGE", |
---|
| 153 | "THETA_IMAGE", |
---|
| 154 | "ELONGATION", |
---|
| 155 | "ELLIPTICITY", |
---|
| 156 | "FWHM_IMAGE", |
---|
| 157 | "ERRX2_IMAGE", |
---|
| 158 | "ERRY2_IMAGE", |
---|
| 159 | "I_GRID", |
---|
| 160 | "J_GRID", |
---|
| 161 | "KNOT_DIST", |
---|
| 162 | "MEAN_DIST_TO_NGBRS"] |
---|
| 163 | |
---|
| 164 | fmts = ["%10.3f ", |
---|
| 165 | "%8.3f ", |
---|
| 166 | "%8.3f ", |
---|
| 167 | "%8.3f ", |
---|
| 168 | "%8.3f ", |
---|
| 169 | "%8.3f ", |
---|
| 170 | "%8.3f ", |
---|
| 171 | "%8.5f ", |
---|
| 172 | "%8.5f ", |
---|
| 173 | "%5d ", |
---|
| 174 | "%5d ", |
---|
| 175 | "%5.3f ", |
---|
| 176 | "%8.3f"] |
---|
| 177 | |
---|
| 178 | for source in cleaned_catalog: |
---|
| 179 | outs = "" |
---|
| 180 | for i in range(len(fields)): |
---|
| 181 | outs += fmts[i] % source[fields[i]] |
---|
| 182 | outs += "\n" |
---|
| 183 | of.write(outs) |
---|
| 184 | |
---|
| 185 | of.close() |
---|
| 186 | |
---|
| 187 | # Génération d'un fichier fits contenant les résultats |
---|
| 188 | |
---|
| 189 | |
---|
| 190 | import pyfits |
---|
| 191 | |
---|
| 192 | #a = zeros((2048,2048)) |
---|
| 193 | # |
---|
| 194 | # patches width in pixel: |
---|
| 195 | ## width = int(modulus(vec_i)) |
---|
| 196 | |
---|
| 197 | ## for i in range(len(cleaned_catalog)): |
---|
| 198 | ## print i |
---|
| 199 | ## source = cleaned_catalog[i] |
---|
| 200 | ## pos_source = array([int(source['X_IMAGE']),int(source['Y_IMAGE'])]) |
---|
| 201 | ## pos_mean_dist = source['MEAN_DIST_TO_NGBRS'] |
---|
| 202 | ## for rel_x in range(-width,width): |
---|
| 203 | ## for rel_y in range(-width,width): |
---|
| 204 | ## rel_vec = array([rel_x,rel_y]) |
---|
| 205 | ## # Forme du patch : carré |
---|
| 206 | ## if ( (dot(rel_vec,vec_i) < vec_i_2/2) and |
---|
| 207 | ## (dot(rel_vec,vec_j) < vec_j_2/2)): |
---|
| 208 | ## (x,y) = pos_source+rel_vec |
---|
| 209 | ## if ( x>0 and x<2048 and y>0 and y<2048): |
---|
| 210 | ## a[x,y] = pos_mean_dist |
---|
| 211 | |
---|
| 212 | |
---|
| 213 | |
---|
| 214 | a = lookup_sources[:,:,3] |
---|
| 215 | |
---|
| 216 | hdu = pyfits.PrimaryHDU(a) |
---|
| 217 | hdulist = pyfits.HDUList([hdu]) |
---|
| 218 | |
---|
| 219 | hdu.writeto(options.outputfits,clobber=True) |
---|
| 220 | |
---|
| 221 | # Facteur d'échelle moyen |
---|
| 222 | |
---|
| 223 | tmp = [] |
---|
| 224 | for i in range(81): |
---|
| 225 | for j in range(81): |
---|
| 226 | if (lookup_sources[i,j,0] > 0): |
---|
| 227 | tmp.append(lookup_sources[i,j,3]) |
---|
| 228 | |
---|
| 229 | print "Moyenne des mesures : ",sum(tmp)/len(tmp) |
---|
| 230 | |
---|
| 231 | # Module des vecteurs unitaires |
---|
| 232 | |
---|
| 233 | print "Module vec_i :",modulus(vec_i) |
---|
| 234 | print "Module vec_j :",modulus(vec_j) |
---|