Mercurial > repos > PredictPDPS / file comparison
comparison: src/ImGenerate.jl
src/ImGenerate.jl
- changeset 24
- 2d9e64235ba7
- parent 22
- a30a8745a83b
- child 25
- c9b06736a477
equal
deleted
inserted
replaced
| 23:1c4b7d1f261f | 24:2d9e64235ba7 |
|---|---|
| 198 while true | 198 while true |
| 199 # Extract subwindow of original image and add noise | 199 # Extract subwindow of original image and add noise |
| 200 b_true = zeros(sz...) | 200 b_true = zeros(sz...) |
| 201 extract_subimage!(b_true, im, v_cumul; threads=true) | 201 extract_subimage!(b_true, im, v_cumul; threads=true) |
| 202 b = b_true .+ params.noise_level.*randn(rng,sz...) | 202 b = b_true .+ params.noise_level.*randn(rng,sz...) |
| 203 v = v_true.*(1.0 .+ params.shake_noise_level.*randn(rng,size(v_true)...)) | 203 v = v_true.*(1.0 .+ zero_factor*params.shake_noise_level.*randn(rng,size(v_true)...)) |
| 204 # Pass data to iteration routine | 204 # Pass data to iteration routine |
| 205 data = OnlineData{DisplacementConstant}(b_true, b, v, v_true, v_cumul) | 205 data = OnlineData{DisplacementConstant}(b_true, b, v, v_true, v_cumul) |
| 206 if !put_unless_closed!(datachannel, data) | 206 if !put_unless_closed!(datachannel, data) |
| 207 return | 207 return |
| 208 end | 208 end |
| 225 | 225 |
| 226 | 226 |
| 227 ######################################################################## | 227 ######################################################################## |
| 228 # PETscan | 228 # PETscan |
| 229 ######################################################################## | 229 ######################################################################## |
| 230 function generate_radon(im, sz, | 230 function generate_sinogram(im, sz, |
| 231 :: Type{DisplacementConstant}, | 231 :: Type{DisplacementConstant}, |
| 232 datachannel :: Channel{PetOnlineData{DisplacementConstant}}, | 232 datachannel :: Channel{PetOnlineData{DisplacementConstant}}, |
| 233 params :: NamedTuple) | 233 params :: NamedTuple) |
| 234 | 234 |
| 235 # Set up counter and zero factor for stabilisation | 235 # Set up counter and zero factor for stabilisation |
| 255 tform = recenter(RotMatrix(theta_cumul), center_point) | 255 tform = recenter(RotMatrix(theta_cumul), center_point) |
| 256 | 256 |
| 257 # Apply the transformation to the image using warp | 257 # Apply the transformation to the image using warp |
| 258 b_true = copy(warp(im, tform, axes(im), fillvalue=Flat())) | 258 b_true = copy(warp(im, tform, axes(im), fillvalue=Flat())) |
| 259 | 259 |
| 260 v = v_true.*(1.0 .+ params.shake_noise_level.*randn(rng,size(v_true)...)) | 260 v = v_true.*(1.0 .+ zero_factor*params.shake_noise_level.*randn(rng,size(v_true)...)) |
| 261 theta = theta_true*(1.0 + params.rotation_noise_level.*randn(rng)) | 261 theta = theta_true*(1.0 + zero_factor*params.rotation_noise_level.*randn(rng)) |
| 262 | 262 |
| 263 # Generate the true and noisy sinograms | 263 # Generate the true and noisy sinograms |
| 264 sinogram_true = zeros(params.radondims...) | 264 sinogram_true = zeros(params.radondims...) |
| 265 sinogram_true .*= params.scale | 265 sinogram_true .*= params.scale |
| 266 radon!(sinogram_true, b_true) | 266 radon!(sinogram_true, b_true) |
| 291 end | 291 end |
| 292 | 292 |
| 293 function imgen_shepplogan_radon(sz) | 293 function imgen_shepplogan_radon(sz) |
| 294 im = convert(Array{Float64},TestImages.shepp_logan(sz[1], highContrast=true)) | 294 im = convert(Array{Float64},TestImages.shepp_logan(sz[1], highContrast=true)) |
| 295 dynrange = maximum(im) | 295 dynrange = maximum(im) |
| 296 return ImGen(curry(generate_radon, im, sz), sz, 1, dynrange, "shepplogan$(sz[1])x$(sz[2])") | 296 return ImGen(curry(generate_sinogram, im, sz), sz, 1, dynrange, "shepplogan$(sz[1])x$(sz[2])") |
| 297 end | 297 end |
| 298 | 298 |
| 299 function imgen_brainphantom_radon(sz) | 299 function imgen_brainphantom_radon(sz) |
| 300 data = matread("src/PET/phantom_slice.mat") | 300 data = matread("src/PET/phantom_slice.mat") |
| 301 im = normalise(imresize(convert(Array{Float64},data["square_data"]),sz)) | 301 im = normalise(imresize(convert(Array{Float64},data["square_data"]),sz)) |
| 302 dynrange = maximum(im) | 302 dynrange = maximum(im) |
| 303 return ImGen(curry(generate_radon, im, sz), sz, 1, dynrange, "brainphantom$(sz[1])x$(sz[2])") | 303 return ImGen(curry(generate_sinogram, im, sz), sz, 1, dynrange, "brainphantom$(sz[1])x$(sz[2])") |
| 304 end | 304 end |
| 305 | 305 |
| 306 normalise = (data) -> data./maximum(data) | 306 normalise = (data) -> data./maximum(data) |
| 307 end # Module | 307 end # Module |
