Lossy Image Compression Attribute
| Tag | (0028,2110) |
|---|---|
| Type | Required (1) |
| Keyword | LossyImageCompression |
| Value Multiplicity | 1 |
| Value Representation | Code String (CS) |
| Example Values |
|
Specifies whether an Image has undergone lossy compression (at a point in its lifetime), or is derived from lossy compressed images.
Enumerated Values:
- 00
-
Image has not been subjected to lossy compression.
- 01
-
Image has been subjected to lossy compression.
Once this value has been set to "01" it shall not be reset.
See Section C.8.20.2.2 and Section C.7.6.1.1.5, “Lossy Image Compression”.
Section C.8.20.2.2
C.8.20.2.2 Lossy Image Compression and Lossy Image Compression Method
If Lossy Image Compression (0028,2110) in any of the source images is "01", the value shall be "01" for the Segmentation Instance.
The process of segmentation itself is defined not to be lossy compression, even though it involves loss. If the Segmentation Instance is encoded using a lossy compression Transfer Syntax, then the value shall be set to "01".
Note
It is not advisable to lossy compress a Segmentation SOP Instance. In particular, a binary segmentation should not be lossy compressed.
Section C.7.6.1.1.5, “Lossy Image Compression”
C.7.6.1.1.5 Lossy Image Compression
The Attribute Lossy Image Compression (0028,2110) conveys that the Image has undergone lossy compression. It provides a means to record that the Image has been compressed (at a point in its lifetime) with a lossy algorithm and changes have been introduced into the pixel data. Once the value has been set to "01", it shall not be reset.
Note
If an image is compressed with a lossy algorithm, the Attribute Lossy Image Compression (0028,2110) is set to "01". Subsequently, if the image is decompressed and transferred in uncompressed format, this Attribute value remains "01".
The value of Lossy Image Compression (0028,2110) in SOP Instances containing multiple frames in which one or more of the frames have undergone lossy compression shall be "01".
Note
It is recommended that the applicable frames be noted in the Attribute Derivation Description (0008,2111).
If an image is originally obtained as a lossy compressed image from the sensor, then Lossy Image Compression (0028,2110) is set to "01" and Value 1 of the Attribute Image Type (0008,0008) shall be set to ORIGINAL.
If an image is a compressed version of another image, Lossy Image Compression (0028,2110) is set to "01", Value 1 of the Attribute Image Type (0008,0008) shall be set to DERIVED, and if the predecessor was a DICOM image, then the Image shall receive a new SOP Instance UID.
Note
-
It is recommended that the approximate compression ratio be provided in the Attribute Derivation Description (0008,2111). Furthermore, it is recommended that Derivation Description (0008,2111) be used to indicate when pixel data changes might affect professional interpretation (see Section C.12.4.1.1).
-
The Attribute Lossy Image Compression (0028,2110) is defined as Type 3 for backward compatibility with existing IODs. It is expected to be required (i.e., defined as Type 1C) for new Image IODs and for existing IODs that undergo a major revision (e.g., a new IOD is specified).
C.7.6.1.1.5.1 Lossy Image Compression Method
Lossy Image Compression Method (0028,2114) may be multi-valued if successive lossy compression steps have been applied; the value order shall correspond to the values of Lossy Image Compression Ratio (0028,2112), if present.
Defined Terms for Lossy Image Compression Method (0028,2114):
- ISO_10918_1
-
JPEG Lossy Compression [ISO/IEC 10918-1]
- ISO_14495_1
-
JPEG-LS Near-lossless Compression [ISO/IEC 14495-1]
- ISO_15444_1
-
JPEG 2000 Irreversible Compression [ISO/IEC 15444-1]
- ISO_15444_15
-
High-Throughput JPEG 2000 Irreversible Compression [ISO/IEC 15444-15]
- ISO_13818_2
-
MPEG2 Compression [ISO/IEC 13818-2]
- ISO_14496_10
-
MPEG-4 AVC/H.264 Compression [ISO/IEC 14496-10]
- ISO_23008_2
-
HEVC/H.265 Lossy Compression [ISO/IEC 23008-2]
C.7.6.1.1.5.2 Lossy Image Compression Ratio
The value of the "compression ratio" is encoded as a numeric value that represents the numerator of an implicit ratio in which the denominator is always one, consistent with the traditional representation in the literature.
Note
For example, a compression ratio of 30:1 would be described with a value of 30.
The value may be an estimate (e.g., the nominal value that is supplied to the compressor), or it may be a measured value (e.g., computed by dividing the uncompressed pixel data size by the size of the compressed bit stream).
Lossy Image Compression Ratio (0028,2112) may be multi-valued if successive lossy compression steps have been applied; if so, the value order shall correspond to the multiple values of Lossy Image Compression Method (0028,2114), if present.
Note
For historical reasons, the lossy compression ratio should also be described in Derivation Description (0008,2111).