mtk平台都有什么测试模式（DCMTK学习笔记2传输语法）

如果是压缩文件，则需要先进行解压操作

官方Demo

DJEncoderRegistration::registerCodecs(); // register JPEG codecs DcmFileFormat fileformat; if (fileformat.loadFile("test.dcm").good()) { DcmDataset *dataset = fileformat.getDataset(); DcmItem *MetaInfo = fileformat.getMetaInfo(); DJ_RPLossless params; // codec parameters, we use the defaults // this causes the lossless JPEG version of the dataset to be created if (dataset->chooseRepresentation(EXS_JPEGProcess14SV1, ¶ms).good() && dataset->canWriteXfer(EXS_JPEGProcess14SV1)) { // store in lossless JPEG format fileformat.saveFile("test_jpeg.dcm", EXS_JPEGProcess14SV1); } } DJEncoderRegistration::cleanup(); // deregister JPEG codecs

压缩的时候示例采用的是默认参数，这个有多个

默认

DJ_RPLossless params;

常用的

D2RepresentationParameter( double compressionRatio = 1, Uint32 compressedSize = 0, OFBool losslessProcess = OFTrue, const OFString &imageComment = "")

这个类有三个参数

压缩某个dicom文件可能会出现这样的错误

Pixel representation cannot be changed

查看源码后得知错误原因可能有

...... if (resultStack.top()->ident() == EVR_PixelData) { DcmPixelData *pixelData = OFstatic_cast(DcmPixelData *, resultStack.top()); if (!pixelData->canChooseRepresentation(repType, repParam)) l_error = EC_CannotChangeRepresentation; pixelStack.push(resultStack); } else l_error = EC_CannotChangeRepresentation; ......

错误类型正是这个

makeOFConditionConst(EC_CannotChangeRepresentation, OFM_dcmdata, 14, OF_error, "Pixel representation cannot be changed" );

DJDecoderRegistration::registerCodecs(); // register JPEG codecs DcmFileFormat fileformat; if (fileformat.loadFile("test_jpeg.dcm").good()) { DcmDataset *dataset = fileformat.getDataset(); // decompress data set if compressed if (dataset->chooseRepresentation(EXS_LittleEndianExplicit, NULL).good() && dataset->canWriteXfer(EXS_LittleEndianExplicit)) { fileformat.saveFile("test_decompressed.dcm", EXS_LittleEndianExplicit); } } DJDecoderRegistration::cleanup(); // deregister JPEG codecs

dcmtk内部初始化时，枚举了所有的压缩类型对应的属性，下面这个是基本数据结构，每个压缩类型对应一个结构体对象

typedef struct { const char *xferID; const char *xferName; E_TransferSyntax xfer; E_ByteOrder byteOrder; E_VRType vrType; E_JPEGEncapsulated encapsulated; Uint32 JPEGProcess8; Uint32 JPEGProcess12; OFBool lossy; OFBool retired; E_StreamCompression streamCompression; } S_XferNames;

DJ_RPLossless params; // codec parameters, we use the defaultsE_TransferSyntax xfer = EXS_JPEGProcess2_4;OFCondition ofResult = dataset->chooseRepresentation(xfer, ¶ms);if (ofResult.good() && dataset->canWriteXfer(xfer)){// store in lossless JPEG formatOFCondition ofResult = fileformat.saveFile("G:/dicomFile/8_test_jpeg.dcm", xfer);}

可以很明显看到图像被压缩的太狠，查看文件大小仅有大概 5KB，原始文件大小288KB

这种方式几乎不会用，因为图像被压缩的太狠，很多关键信息都丢失了

int nQuality = 50; // this causes the lossless JPEG version of the dataset to be created D2RepresentationParameter rpp( nQuality*0.01, OFstatic_cast(Uint32, 1), OFFalse, ""); OFCondition ofResult = dataset->chooseRepresentation(xfer, &rpp);

这种情况下，压缩后文件大小为 33KB左右，显示效果比上一个好点

一般默认可以采用 EXS_JPEGProcess2_4进行压缩，如果图像包含Overlay层，则要使用 EXS_JPEG2000协议，对于常见的CT,MR,NM,PET等，如果图像宽高小于 1024 情况下可以采用 EXS_JPEGProcess14SV1协议