mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-11-30 23:49:05 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Replace malloc / free for STATISTICS

Browse Source 
Signed-off-by: Stefan Weil <sw@weilnetz.de>
This commit is contained in:
Stefan Weil 2021-03-27 23:22:28 +01:00
parent 57d3a1eb99
commit 441f74c1e6
1 changed files with 16 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

61
src/classify/cluster.cpp
View File

@ -1263,10 +1263,12 @@ using ClusterPair = tesseract::KDPairInc<float, TEMPCLUSTER *>;
using ClusterHeap = tesseract::GenericHeap<ClusterPair>; using ClusterHeap = tesseract::GenericHeap<ClusterPair>;
struct STATISTICS { struct STATISTICS {
float AvgVariance; STATISTICS(size_t n) : CoVariance(n * n), Min(n), Max(n) {
float *CoVariance; }
float *Min; // largest negative distance from the mean float AvgVariance = 1.0f;
float *Max; // largest positive distance from the mean std::vector<float> CoVariance;
std::vector<float> Min; // largest negative distance from the mean
std::vector<float> Max; // largest positive distance from the mean
}; };
struct BUCKETS { struct BUCKETS {
@ -1405,8 +1407,6 @@ static uint16_t UniformBucket(PARAM_DESC *ParamDesc, float x, float Mean, float
static bool DistributionOK(BUCKETS *Buckets); static bool DistributionOK(BUCKETS *Buckets);
static void FreeStatistics(STATISTICS *Statistics);
static void FreeBuckets(BUCKETS *Buckets); static void FreeBuckets(BUCKETS *Buckets);
static uint16_t DegreesOfFreedom(DISTRIBUTION Distribution, uint16_t HistogramBuckets); static uint16_t DegreesOfFreedom(DISTRIBUTION Distribution, uint16_t HistogramBuckets);
@ -1955,7 +1955,6 @@ static void ComputePrototypes(CLUSTERER *Clusterer, CLUSTERCONFIG *Config) {
* @return Pointer to new prototype or nullptr * @return Pointer to new prototype or nullptr
*/ */
static PROTOTYPE *MakePrototype(CLUSTERER *Clusterer, CLUSTERCONFIG *Config, CLUSTER *Cluster) { static PROTOTYPE *MakePrototype(CLUSTERER *Clusterer, CLUSTERCONFIG *Config, CLUSTER *Cluster) {
STATISTICS *Statistics;
PROTOTYPE *Proto; PROTOTYPE *Proto;
BUCKETS *Buckets; BUCKETS *Buckets;
@ -1965,7 +1964,7 @@ static PROTOTYPE *MakePrototype(CLUSTERER *Clusterer, CLUSTERCONFIG *Config, CLU
} }
// compute the covariance matrix and ranges for the cluster // compute the covariance matrix and ranges for the cluster
Statistics = ComputeStatistics(Clusterer->SampleSize, Clusterer->ParamDesc, Cluster); auto Statistics = ComputeStatistics(Clusterer->SampleSize, Clusterer->ParamDesc, Cluster);
// check for degenerate clusters which need not be analyzed further // check for degenerate clusters which need not be analyzed further
// note that the MinSamples test assumes that all clusters with multiple // note that the MinSamples test assumes that all clusters with multiple
@ -1973,20 +1972,20 @@ static PROTOTYPE *MakePrototype(CLUSTERER *Clusterer, CLUSTERCONFIG *Config, CLU
Proto = MakeDegenerateProto(Clusterer->SampleSize, Cluster, Statistics, Config->ProtoStyle, Proto = MakeDegenerateProto(Clusterer->SampleSize, Cluster, Statistics, Config->ProtoStyle,
static_cast<int32_t>(Config->MinSamples * Clusterer->NumChar)); static_cast<int32_t>(Config->MinSamples * Clusterer->NumChar));
if (Proto != nullptr) { if (Proto != nullptr) {
FreeStatistics(Statistics); delete Statistics;
return Proto; return Proto;
} }
// check to ensure that all dimensions are independent // check to ensure that all dimensions are independent
if (!Independent(Clusterer->ParamDesc, Clusterer->SampleSize, Statistics->CoVariance, if (!Independent(Clusterer->ParamDesc, Clusterer->SampleSize, &Statistics->CoVariance[0],
Config->Independence)) { Config->Independence)) {
FreeStatistics(Statistics); delete Statistics;
return nullptr; return nullptr;
} }
if (HOTELLING && Config->ProtoStyle == elliptical) { if (HOTELLING && Config->ProtoStyle == elliptical) {
Proto = TestEllipticalProto(Clusterer, Config, Cluster, Statistics); Proto = TestEllipticalProto(Clusterer, Config, Cluster, Statistics);
if (Proto != nullptr) { if (Proto != nullptr) {
FreeStatistics(Statistics); delete Statistics;
return Proto; return Proto;
} }
} }
@ -2017,7 +2016,7 @@ static PROTOTYPE *MakePrototype(CLUSTERER *Clusterer, CLUSTERCONFIG *Config, CLU
Proto = MakeMixedProto(Clusterer, Cluster, Statistics, Buckets, Config->Confidence); Proto = MakeMixedProto(Clusterer, Cluster, Statistics, Buckets, Config->Confidence);
break; break;
} }
FreeStatistics(Statistics); delete Statistics;
return Proto; return Proto;
} // MakePrototype } // MakePrototype
@ -2368,33 +2367,18 @@ static void MakeDimUniform(uint16_t i, PROTOTYPE *Proto, STATISTICS *Statistics)
* @return Pointer to new data structure containing statistics * @return Pointer to new data structure containing statistics
*/ */
static STATISTICS *ComputeStatistics(int16_t N, PARAM_DESC ParamDesc[], CLUSTER *Cluster) { static STATISTICS *ComputeStatistics(int16_t N, PARAM_DESC ParamDesc[], CLUSTER *Cluster) {
STATISTICS *Statistics;
int i, j; int i, j;
float *CoVariance;
float *Distance; float *Distance;
LIST SearchState; LIST SearchState;
SAMPLE *Sample; SAMPLE *Sample;
uint32_t SampleCountAdjustedForBias; uint32_t SampleCountAdjustedForBias;
// allocate memory to hold the statistics results // allocate memory to hold the statistics results
Statistics = static_cast<STATISTICS *>(malloc(sizeof(STATISTICS))); auto Statistics = new STATISTICS(N);
Statistics->CoVariance = static_cast<float *>(malloc(sizeof(float) * N * N));
Statistics->Min = static_cast<float *>(malloc(N * sizeof(float)));
Statistics->Max = static_cast<float *>(malloc(N * sizeof(float)));
// allocate temporary memory to hold the sample to mean distances // allocate temporary memory to hold the sample to mean distances
Distance = static_cast<float *>(malloc(N * sizeof(float))); Distance = static_cast<float *>(malloc(N * sizeof(float)));
// initialize the statistics
Statistics->AvgVariance = 1.0;
CoVariance = Statistics->CoVariance;
for (i = 0; i < N; i++) {
Statistics->Min[i] = 0.0;
Statistics->Max[i] = 0.0;
for (j = 0; j < N; j++, CoVariance++) {
*CoVariance = 0;
}
}
// find each sample in the cluster and merge it into the statistics // find each sample in the cluster and merge it into the statistics
InitSampleSearch(SearchState, Cluster); InitSampleSearch(SearchState, Cluster);
while ((Sample = NextSample(&SearchState)) != nullptr) { while ((Sample = NextSample(&SearchState)) != nullptr) {
@ -2415,7 +2399,7 @@ static STATISTICS *ComputeStatistics(int16_t N, PARAM_DESC ParamDesc[], CLUSTER
Statistics->Max[i] = Distance[i]; Statistics->Max[i] = Distance[i];
} }
} }
CoVariance = Statistics->CoVariance; auto CoVariance = &Statistics->CoVariance[0];
for (i = 0; i < N; i++) { for (i = 0; i < N; i++) {
for (j = 0; j < N; j++, CoVariance++) { for (j = 0; j < N; j++, CoVariance++) {
*CoVariance += Distance[i] * Distance[j]; *CoVariance += Distance[i] * Distance[j];
@ -2431,7 +2415,7 @@ static STATISTICS *ComputeStatistics(int16_t N, PARAM_DESC ParamDesc[], CLUSTER
} else { } else {
SampleCountAdjustedForBias = 1; SampleCountAdjustedForBias = 1;
} }
CoVariance = Statistics->CoVariance; auto CoVariance = &Statistics->CoVariance[0];
for (i = 0; i < N; i++) { for (i = 0; i < N; i++) {
for (j = 0; j < N; j++, CoVariance++) { for (j = 0; j < N; j++, CoVariance++) {
*CoVariance /= SampleCountAdjustedForBias; *CoVariance /= SampleCountAdjustedForBias;
@ -2493,7 +2477,6 @@ static PROTOTYPE *NewSphericalProto(uint16_t N, CLUSTER *Cluster, STATISTICS *St
*/ */
static PROTOTYPE *NewEllipticalProto(int16_t N, CLUSTER *Cluster, STATISTICS *Statistics) { static PROTOTYPE *NewEllipticalProto(int16_t N, CLUSTER *Cluster, STATISTICS *Statistics) {
PROTOTYPE *Proto; PROTOTYPE *Proto;
float *CoVariance;
int i; int i;
Proto = NewSimpleProto(N, Cluster); Proto = NewSimpleProto(N, Cluster);
@ -2501,7 +2484,7 @@ static PROTOTYPE *NewEllipticalProto(int16_t N, CLUSTER *Cluster, STATISTICS *St
Proto->Magnitude.Elliptical = static_cast<float *>(malloc(N * sizeof(float))); Proto->Magnitude.Elliptical = static_cast<float *>(malloc(N * sizeof(float)));
Proto->Weight.Elliptical = static_cast<float *>(malloc(N * sizeof(float))); Proto->Weight.Elliptical = static_cast<float *>(malloc(N * sizeof(float)));
CoVariance = Statistics->CoVariance; auto CoVariance = &Statistics->CoVariance[0];
Proto->TotalMagnitude = 1.0; Proto->TotalMagnitude = 1.0;
for (i = 0; i < N; i++, CoVariance += N + 1) { for (i = 0; i < N; i++, CoVariance += N + 1) {
Proto->Variance.Elliptical[i] = *CoVariance; Proto->Variance.Elliptical[i] = *CoVariance;
@ -3053,18 +3036,6 @@ static bool DistributionOK(BUCKETS *Buckets) {
} }
} // DistributionOK } // DistributionOK
/**
* This routine frees the memory used by the statistics
* data structure.
* @param Statistics pointer to data structure to be freed
*/
static void FreeStatistics(STATISTICS *Statistics) {
free(Statistics->CoVariance);
free(Statistics->Min);
free(Statistics->Max);
free(Statistics);
} // FreeStatistics
/** /**
* This routine properly frees the memory used by a BUCKETS. * This routine properly frees the memory used by a BUCKETS.
* *