protected static void read_header_and_symbols(IGenericFst fst, BinaryReader reader)
{
if (read_int32_LE(reader) != OPENFST_MAGIC)
throw new Exception("invalid magic number");
read_magic_string(reader, "vector");
read_magic_string(reader, "standard");
int version = read_int32_LE(reader);
if (version < MIN_VERSION)
throw new Exception("file has too old version");
int flags = read_int32_LE(reader);
read_int64_LE(reader); // properties
Int64 start = read_int64_LE(reader);
Int64 nstates = read_int64_LE(reader);
if (nstates < 0)
return; // to prevent creating 2^31 nodes in case of sudden EOF
fst.Clear();
for (int i = 0; i < nstates; i++)
fst.NewState();
fst.SetStart((int)start);
read_int64_LE(reader); // narcs
if ((flags & FLAG_HAS_ISYMBOLS) > 0)
skip_symbol_table(reader);
if ((flags & FLAG_HAS_OSYMBOLS) > 0)
skip_symbol_table(reader);
}
public static void a_star_backwards(Floatarray costs_for_all_nodes, IGenericFst fst)
{
IGenericFst reverse = FstFactory.MakeOcroFST();
FstUtil.fst_copy_reverse(reverse, fst, true); // creates an extra vertex
AStarSearch a = new AStarSearch(reverse);
a.Loop();
costs_for_all_nodes.Copy(a.g);
costs_for_all_nodes.Pop(); // remove the extra vertex
}
int n; // the number of nodes; also the virtual accept index
#endregion Fields
#region Constructors
public AStarSearch(IGenericFst fst)
{
this.fst = fst;
this.accepted_from = -1;
this.heap = new Heap(fst.nStates() + 1);
this.n = fst.nStates();
this.came_from = new Intarray(n);
this.came_from.Fill(-1);
this.g = new Floatarray(n);
// insert the start node
int s = fst.GetStart();
g[s] = 0;
came_from[s] = s;
heap.Push(s, Convert.ToSingle(Heuristic(s)));
}
public CompositionFstImpl(IGenericFst l1, IGenericFst l2,
int o_s, int o_f)
{
override_start = o_s;
override_finish = o_f;
if (l1.nStates() == 0)
throw new Exception("CHECK_ARG: l1->nStates() > 0");
if (l2.nStates() == 0)
throw new Exception("CHECK_ARG: l2->nStates() > 0");
// this should be here, not in the initializers.
// (otherwise if CHECKs throw an exception, bad things happen)
this.l1 = l1;
this.l2 = l2;
}
/// <summary>
/// Copy one FST to another, preserving only lowest-cost arcs.
/// This is useful for visualization.
/// </summary>
/// <param name="dst">The destination. Will be cleared before copying.</param>
/// <param name="src">The FST to copy.</param>
public static void fst_copy_best_arcs_only(IGenericFst dst, IGenericFst src)
{
dst.Clear();
int n = src.nStates();
for (int i = 0; i < n; i++)
dst.NewState();
dst.SetStart(src.GetStart());
for(int i = 0; i < n; i++)
{
dst.SetAccept(i, src.GetAcceptCost(i));
Intarray targets = new Intarray(), outputs = new Intarray(), inputs = new Intarray();
Floatarray costs = new Floatarray();
src.Arcs(inputs, targets, outputs, costs, i);
int inlen = inputs.Length();
if (inlen != targets.Length())
throw new Exception("ASSERT: inputs.length() == targets.length()");
if (inlen != outputs.Length())
throw new Exception("ASSERT: inputs.length() == outputs.length()");
if (inlen != costs.Length())
throw new Exception("ASSERT: inputs.length() == costs.length()");
Dictionary< int, int > hash = new Dictionary<int,int>();
for(int j = 0; j < n; j++) {
int t = targets[j];
int best_so_far = -1;
if (hash.ContainsKey(t))
best_so_far = hash[t];
if(best_so_far == -1 || costs[j] < costs[best_so_far])
hash[t] = j;
}
Intarray keys = new Intarray();
//hash.keys(keys);
keys.Clear();
foreach (int key in hash.Keys)
{
keys.Push(key);
}
for(int k = 0; k < keys.Length(); k++) {
int j = hash[keys[k]];
dst.AddTransition(i, targets[j], outputs[j], costs[j], inputs[j]);
}
}
}
/// <summary>
/// Copy one FST to another.
/// </summary>
/// <param name="dst">The destination. Will be cleared before copying.</param>
/// <param name="src">The FST to copy.</param>
public static void fst_copy(IGenericFst dst, IGenericFst src)
{
dst.Clear();
int n = src.nStates();
for (int i = 0; i < n; i++)
dst.NewState();
dst.SetStart(src.GetStart());
for (int i = 0; i < n; i++)
{
dst.SetAccept(i, src.GetAcceptCost(i));
Intarray targets = new Intarray(), outputs = new Intarray(), inputs = new Intarray();
Floatarray costs = new Floatarray();
src.Arcs(inputs, targets, outputs, costs, i);
int inlen = inputs.Length();
if (inlen != targets.Length())
throw new Exception("ASSERT: inputs.length() == targets.length()");
if (inlen != outputs.Length())
throw new Exception("ASSERT: inputs.length() == outputs.length()");
if (inlen != costs.Length())
throw new Exception("ASSERT: inputs.length() == costs.length()");
for (int j = 0; j < inputs.Length(); j++)
dst.AddTransition(i, targets.At1d(j), outputs.At1d(j), costs.At1d(j), inputs.At1d(j));
}
}
/// <summary>
/// recognize a line with or without a given segmentation
/// if useit is set to true, the given segmentation is just displayed in loggers, but not used,
/// the segmenter computes the segmentation and the recognition uses its output
/// if useit is set to false, the segmenter is still launched for the loggers, but the given
/// segmentation is really used for the recognition
/// </summary>
public virtual double RecognizeLineSeg(IGenericFst result, Intarray segmentation, Bytearray image)
{
return this.RecognizeLine(segmentation, result, image);
}
/// <summary> /// Recognize a text line and return a lattice representing /// the recognition alternatives. /// </summary> public abstract double RecognizeLine(IGenericFst result, Bytearray image);
public override IGenericFst Move2()
{
IGenericFst result = l2;
l2 = null;
return result;
}
protected static void read_node(BinaryReader reader, IGenericFst fst, int index)
{
fst.SetAccept(index, read_float(reader));
Int64 narcs = read_int64_LE(reader);
for (int i = 0; i < narcs; i++)
{
int input = read_int32_LE(reader);
int output = read_int32_LE(reader);
float cost = read_float(reader);
int target = read_int32_LE(reader);
fst.AddTransition(index, target, output, cost, input);
}
}
public static void fst_write(BinaryWriter writer, IGenericFst fst)
{
write_header_and_symbols(writer, fst);
for (int i = 0; i < fst.nStates(); i++)
write_node(writer, fst, i);
}
protected static void write_node(BinaryWriter writer, IGenericFst fst, int index)
{
Intarray inputs = new Intarray();
Intarray targets = new Intarray();
Intarray outputs = new Intarray();
Floatarray costs = new Floatarray();
fst.Arcs(inputs, targets, outputs, costs, index);
int narcs = targets.Length();
write_float(writer, fst.GetAcceptCost(index));
write_int64_LE(writer, narcs);
for (int i = 0; i < narcs; i++)
{
write_int32_LE(writer, inputs[i]);
write_int32_LE(writer, outputs[i]);
write_float(writer, costs[i]);
write_int32_LE(writer, targets[i]);
}
}
/// <summary>
/// Reverse the FST's arcs, adding a new start vertex (former accept).
/// </summary>
public static void fst_copy_reverse(IGenericFst dst, IGenericFst src, bool no_accept = false)
{
dst.Clear();
int n = src.nStates();
for (int i = 0; i <= n; i++)
dst.NewState();
if (!no_accept)
dst.SetAccept(src.GetStart());
dst.SetStart(n);
for (int i = 0; i < n; i++)
{
dst.AddTransition(n, i, 0, src.GetAcceptCost(i), 0);
Intarray targets = new Intarray(), outputs = new Intarray(), inputs = new Intarray();
Floatarray costs = new Floatarray();
src.Arcs(inputs, targets, outputs, costs, i);
if (inputs.Length() != targets.Length())
throw new Exception("ASSERT: inputs.length() == targets.length()");
if (inputs.Length() != outputs.Length())
throw new Exception("ASSERT: inputs.length() == outputs.length()");
if (inputs.Length() != costs.Length())
throw new Exception("ASSERT: inputs.length() == costs.length()");
for (int j = 0; j < inputs.Length(); j++)
dst.AddTransition(targets.At1d(j), i, outputs.At1d(j), costs.At1d(j), inputs.At1d(j));
}
}
/// <summary> /// Recognize a text line and return a lattice representing /// the recognition alternatives. /// </summary> public abstract double RecognizeLine(IGenericFst result, Bytearray image);
/// <summary>
/// Align a lattice with a transcription.
/// </summary>
/// <param name="chars">Characters along the best path.
/// Currently, every character in chars must have a corresponding
/// region in seg and the characters must be in reading order.
/// Eventually, chars may contain characters (e.g., spaces) that
/// do not correspond to any region. Note that chars may not
/// correspond to any string allowed/suggested by the transcription.</param>
/// <param name="seg">Aligned segmentation, colors correspond to chars (starting at 1)</param>
/// <param name="costs">Costs corresponding to chars</param>
/// <param name="image">Input grayscale image</param>
/// <param name="transcription">The "ground truth" lattice to align</param>
public virtual void Align(string chars, Intarray seg, Floatarray costs,
Bytearray image, IGenericFst transcription)
{
throw new NotImplementedException("IRecognizeLine:Align: unimplemented");
}
/// <summary>
/// recognize a line with or without a given segmentation
/// if useit is set to true, the given segmentation is just displayed in loggers, but not used,
/// the segmenter computes the segmentation and the recognition uses its output
/// if useit is set to false, the segmenter is still launched for the loggers, but the given
/// segmentation is really used for the recognition
/// </summary>
public virtual double RecognizeLineSeg(IGenericFst result, Intarray segmentation, Bytearray image)
{
return(this.RecognizeLine(segmentation, result, image));
}
/// <summary>
/// This is a weird, optional method that exposes character segmentation
/// for those line recognizers that have it segmentation contains colored pixels,
/// and a transition in the transducer of the form * --- 1/eps --> * --- 2/a --> *
/// means that pixels with color 1 and 2 together form the letter "a"
/// </summary>
public virtual double RecognizeLine(Intarray segmentation, IGenericFst result, Bytearray image)
{
throw new NotImplementedException("IRecognizeLine:RecognizeLine: unimplemented");
}
public void GetLattice(IGenericFst fst, int page, int line, string variant = null)
{
string s = PathFile(page, line, variant, "fst");
fst.Load(s);
}
/// <summary>
/// Compose two FSTs.
/// This function copies the composition of two given FSTs.
/// That causes expansion (storing all arcs explicitly).
/// </summary>
public static void fst_expand_composition(IGenericFst outf, OcroFST f1, OcroFST f2)
{
CompositionFst composition = FstFactory.MakeCompositionFst(f1, f2);
try
{
fst_copy(outf, composition);
}
catch (Exception ex)
{
composition.Move1();
composition.Move2();
throw ex;
}
composition.Move1();
composition.Move2();
}
protected static void write_header_and_symbols(BinaryWriter writer, IGenericFst fst)
{
write_int32_LE(writer, OPENFST_MAGIC);
write_string(writer, "vector");
write_string(writer, "standard");
write_int32_LE(writer, MIN_VERSION);
write_int32_LE(writer, /* flags: */ 0);
write_int64_LE(writer, PROPERTIES);
write_int64_LE(writer, fst.GetStart());
write_int64_LE(writer, fst.nStates());
write_int64_LE(writer, /* narcs (seems to be unused): */ 0L);
}
public static void fst_line(IGenericFst fst, string s)
{
int n = s.Length;
Intarray inputs = new Intarray(n);
for(int j = 0; j < n; j++)
inputs[j] = (int)s[j];
Floatarray costs = new Floatarray(n);
costs.Fill(0f);
fst.SetString(s, costs, inputs);
}
protected static readonly Int64 PROPERTIES = 3; // expanded, mutable
#endregion Fields
#region Methods
public static void fst_read(IGenericFst fst, BinaryReader reader)
{
read_header_and_symbols(fst, reader);
for (int i = 0; i < fst.nStates(); i++)
read_node(reader, fst, i);
}
internal static bool a_star2_internal(Intarray inputs,
Intarray vertices1,
Intarray vertices2,
Intarray outputs,
Floatarray costs,
IGenericFst fst1,
IGenericFst fst2,
Floatarray g1,
Floatarray g2,
CompositionFst composition)
{
Intarray vertices = new Intarray();
AStarCompositionSearch a = new AStarCompositionSearch(g1, g2, composition);
if (!a.Loop())
return false;
if (!a.reconstruct_vertices(vertices))
return false;
a.reconstruct_edges(inputs, outputs, costs, vertices);
composition.SplitIndices(vertices1, vertices2, vertices);
return true;
}
/// <summary>
/// This is a weird, optional method that exposes character segmentation
/// for those line recognizers that have it segmentation contains colored pixels,
/// and a transition in the transducer of the form * --- 1/eps --> * --- 2/a --> *
/// means that pixels with color 1 and 2 together form the letter "a"
/// </summary>
public virtual double RecognizeLine(Intarray segmentation, IGenericFst result, Bytearray image)
{
throw new NotImplementedException("IRecognizeLine:RecognizeLine: unimplemented");
}
/// <summary> /// Extract the lattice corresponding to the classifications /// stored in the Grouper. /// </summary> public abstract void GetLattice(IGenericFst fst);
public override IGenericFst Move1()
{
IGenericFst result = l1;
l1 = null;
return result;
}
/// <summary>
/// This is a weird, optional method that exposes character segmentation
/// for those line recognizers that have it segmentation contains colored pixels,
/// and a transition in the transducer of the form * --- 1/eps --> * --- 2/a --> *
/// means that pixels with color 1 and 2 together form the letter "a"
/// </summary>
public override double RecognizeLine(Intarray segmentation_, IGenericFst result, Bytearray image_)
{
double rate = 0.0;
CHECK_ARG(image_.Dim(1) < PGeti("maxheight"),
String.Format("input line too high ({0} x {1})", image_.Dim(0), image_.Dim(1)));
CHECK_ARG(image_.Dim(1) * 1.0 / image_.Dim(0) < PGetf("maxaspect"),
String.Format("input line has bad aspect ratio ({0} x {1})", image_.Dim(0), image_.Dim(1)));
bool use_reject = PGetb("use_reject") && !DisableJunk;
//Console.WriteLine("IMG: imin:{0} imax:{1}", NarrayUtil.ArgMin(image_), NarrayUtil.ArgMax(image_));
Bytearray image = new Bytearray();
image.Copy(image_);
SetLine(image_);
if (PGeti("invert") > 0)
NarrayUtil.Sub(NarrayUtil.Max(image), image);
segmentation_.Copy(segmentation);
Bytearray available = new Bytearray();
Floatarray cp = new Floatarray();
Floatarray ccosts = new Floatarray();
Floatarray props = new Floatarray();
OutputVector p = new OutputVector();
int ncomponents = grouper.Object.Length();
int minclass = PGeti("minclass");
float minprob = PGetf("minprob");
float space_yes = PGetf("space_yes");
float space_no = PGetf("space_no");
float maxcost = PGetf("maxcost");
// compute priors if possible; fall back on
// using no priors if no counts are available
Floatarray priors = new Floatarray();
bool use_priors = PGeti("use_priors") > 0;
if (use_priors)
{
if (counts.Length() > 0)
{
priors.Copy(counts);
priors /= NarrayUtil.Sum(priors);
}
else
{
if (!counts_warned)
Global.Debugf("warn", "use_priors specified but priors unavailable (old model)");
use_priors = false;
counts_warned = true;
}
}
EstimateSpaceSize();
for (int i = 0; i < ncomponents; i++)
{
Rect b;
Bytearray mask = new Bytearray();
grouper.Object.GetMask(out b, ref mask, i, 0);
Bytearray cv = new Bytearray();
grouper.Object.ExtractWithMask(cv, mask, image, i, 0);
//ImgIo.write_image_gray("extrmask_image.png", cv);
Floatarray v = new Floatarray();
v.Copy(cv);
v /= 255.0f;
float ccost = classifier.Object.XOutputs(p, v);
if (use_reject && classifier.Object.HigherOutputIsBetter)
{
ccost = 0;
float total = p.Sum();
if (total > 1e-11f)
{
//p /= total;
}
else
p.Values.Fill(0.0f);
}
int count = 0;
Global.Debugf("dcost", "output {0}", p.Keys.Length());
for (int index = 0; index < p.Keys.Length(); index++)
{
int j = p.Keys[index];
if (j < minclass) continue;
if (j == reject_class) continue;
float value = p.Values[index];
if (value <= 0.0f) continue;
if (value < minprob) continue;
float pcost = classifier.Object.HigherOutputIsBetter ? (float)-Math.Log(value) : value;
Global.Debugf("dcost", "{0} {1} {2}", j, pcost + ccost, (j > 32 ? (char)j : '_'));
float total_cost = pcost + ccost;
if (total_cost < maxcost)
{
if (use_priors)
{
total_cost -= (float)-Math.Log(priors[j]);
}
grouper.Object.SetClass(i, j, total_cost);
count++;
}
}
Global.Debugf("dcost", "");
if (count == 0)
{
float xheight = 10.0f;
if (b.Height() < xheight / 2 && b.Width() < xheight / 2)
{
grouper.Object.SetClass(i, (int)'~', high_cost / 2);
}
else
{
grouper.Object.SetClass(i, (int)'#', (b.Width() / xheight) * high_cost);
}
}
if (grouper.Object.PixelSpace(i) > space_threshold)
{
Global.Debugf("spaces", "space {0}", grouper.Object.PixelSpace(i));
grouper.Object.SetSpaceCost(i, space_yes, space_no);
}
}
grouper.Object.GetLattice(result);
return rate;
}
/// <summary>
/// Randomly sample an FST, assuming any input.
/// </summary>
/// <param name="result">The array of output symbols, excluding epsilons.</param>
/// <param name="fst">The FST.</param>
/// <param name="max">The maximum length of the result.</param>
/// <returns>total cost</returns>
public static double fst_sample(Intarray result, IGenericFst fst, int max=1000)
{
double total_cost = 0;
int current = fst.GetStart();
for (int counter = 0; counter < max; counter++)
{
Intarray inputs = new Intarray();
Intarray outputs = new Intarray();
Intarray targets = new Intarray();
Floatarray costs = new Floatarray();
fst.Arcs(inputs, targets, outputs, costs, current);
// now we need to deal with the costs uniformly, so:
costs.Push(fst.GetAcceptCost(current));
int choice = sample_by_costs(costs);
if (choice == costs.Length() - 1)
break;
result.Push(outputs[choice]);
total_cost += costs[choice];
current = targets[choice];
}
return total_cost + fst.GetAcceptCost(current);
}
/// <summary>
/// Align a lattice with a transcription.
/// </summary>
/// <param name="chars">Characters along the best path.
/// Currently, every character in chars must have a corresponding
/// region in seg and the characters must be in reading order.
/// Eventually, chars may contain characters (e.g., spaces) that
/// do not correspond to any region. Note that chars may not
/// correspond to any string allowed/suggested by the transcription.</param>
/// <param name="seg">Aligned segmentation, colors correspond to chars (starting at 1)</param>
/// <param name="costs">Costs corresponding to chars</param>
/// <param name="image">Input grayscale image</param>
/// <param name="transcription">The "ground truth" lattice to align</param>
public virtual void Align(string chars, Intarray seg, Floatarray costs,
Bytearray image, IGenericFst transcription)
{
throw new NotImplementedException("IRecognizeLine:Align: unimplemented");
}
/// <summary>
/// Randomly sample an FST, assuming any input.
/// </summary>
public static double fst_sample(out string result, IGenericFst fst, int max)
{
Intarray tmp = new Intarray();
double cost = fst_sample(tmp, fst, max);
remove_epsilons(out result, tmp);
return cost;
}
/// <summary>
/// Make an in-place Kleene closure of the FST.
/// </summary>
public static void fst_star(IGenericFst fst)
{
int s = fst.GetStart();
fst.SetAccept(s);
for (int i = 0; i < fst.nStates(); i++)
{
double c = fst.GetAcceptCost(i);
if (c < 1e37)
fst.AddTransition(i, s, 0, (float)c, 0);
}
}
/// <summary>
/// Recognize a text line and return a lattice representing
/// the recognition alternatives.
/// </summary>
public override double RecognizeLine(IGenericFst result, Bytearray image)
{
Intarray segmentation_ = new Intarray();
return RecognizeLine(segmentation_, result, image);
}
/// <summary>
/// Make a Kleene closure.
/// </summary>
public static void fst_star(IGenericFst result, IGenericFst fst)
{
fst_copy(result, fst);
fst_star(result);
}
/// <summary> /// Extract the lattice corresponding to the classifications /// stored in the Grouper. /// </summary> public abstract void GetLattice(IGenericFst fst);
public void PutLattice(IGenericFst fst, int page, int line, string variant = null)
{
string s = PathFile(page, line, variant, "fst");
fst.Save(s);
}
