G06V 10/20
Definition
Diese Klassifikationsstelle umfasst:(Für diese Definition ist die deutsche Übersetzung noch nicht abgeschlossen)
Any kind of processing of acquired image or video data before the steps of feature extraction and recognition; devices configured to perform this processing.
Processing to prepare an image for feature extraction.
Processing to enhance image quality with the intent to emphasise structures in the image, which inform the automated recognition of objects or categories of objects.
Processing to attenuate or discard elements of the image, which are unlikely to be useful for the pattern recognition process.
Processing converts image to a standard format suitable for feature extraction and pattern recognition routines.
Notes – other classification places
Specific aspects of pre-processing are covered by the subgroups of group G06V 10/20; they particularly relate to aspects such as:
- Processes or devices for identifying regions of the image, which should be subjected to the pattern recognition process, or which are likely to contain image information that is relevant for an object recognition task - covered by group G06V 10/22;
- Correcting wrongly oriented images (e.g. changing the orientation from an erroneous portrait mode to landscape mode), compensating for the pose change of the object by performing affine transformations (translation, scaling, homothety, similarity, reflection, rotation, shear mapping, and compositions of them in any combination and sequence), or for correcting geometrical distortions induced by the image capturing - covered by group G06V 10/24;
- Determination of a bounding box containing the pattern of interest, processing within a region-of-interest [ROI] or volume-of-interest [VOI] to emphasise the pattern for recognition – covered by group G06V 10/25;
- Devices or processes for separating a candidate object from other, non-interesting image regions or the background; image segmentation to the extent that it is adapted to support a subsequent recognition step - covered by group G06V 10/26;
- Adjusting the bit depth, e.g. conversion to black-and-white images, and setting thresholds therefor, e.g. by analysis of the histogram of the image grey levels; Converting the image data to a predetermined numerical range, e.g. by scaling pixel values – covered by group G06V 10/28;
- Techniques for improving the signal-to-noise (SNR) ratio or denoising the image for the purpose of improving the recognition – covered by group G06V 10/30;
- Adjusting the size or the resolution of the image to a standard format, e.g. by scaling; adjusting the size of the detected object to a certain format – covered by group G06V 10/32;
- Smoothing or thinning to obtain an alternative, less complex representation of the pattern; applying morphological operators (e.g. morphological dilation, erosion, opening or closing) for filling in gaps or merging elements, with the aim of emphasising the structures relevant for recognition; Skeleton extraction for characterising the shape of a pattern – covered by group G06V 10/34;
- Enhancing the contrast by convolving the image with a filter mask or by applying a non-linear operator to local image patches – covered by group G06V 10/36.
Examples
Alignment of the image of a face by affine transformations to obtain a pose-invariant image
Beziehungen zu anderen Klassifikationsstellen
Different image preprocessing in general are covered in groups as it follows:
- G06T 3/00 when geometric image transformations (e.g. image rotation) are involved;
- G06T 5/00 when image enhancement or restoration (e.g. denoising) is performed.
Querverweise
Nichteinschränkende Querverweise in anwendungsorientierte Klassifikationsstellen
Recognising scenes; Scene-specific elements
| G06V 20/00 |
Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
| G06V 30/00 |
Image or video recognition or understanding of human-related, animal-related or biometric patterns in image or video data
| G06V 40/00 |
Informative Querverweise
Filter operations to reveal edges, corners, or other image features, which are used to characterise objects
| G06V 10/44 |
Image enhancement or restoration
| G06T 5/00 |
Image segmentation
| G06T 7/10 |
Morphological operators for image segmentation
| G06T 7/155 |
Glossar
DCT
| discrete cosine transform
|
FFT
| fast Fourier transform
|
FOV
| field of view, the region of the environment that an image sensor observes.
|
ROI
| region of interest, an image patch that is likely to contain relevant information.
|
skeletonisation
| process of shrinking a shape to a connected sequence of lines, which are equidistant to the boundaries of the shape.
|
SNR
| signal-to-noise ratio
|
VOI
| volume of interest, a cuboid that encloses three-dimensional data points that are likely to represent relevant information.
|
G06V 10/20
Definition Statement
This place covers:Any kind of processing of acquired image or video data before the steps of feature extraction and recognition; devices configured to perform this processing.
Processing to prepare an image for feature extraction.
Processing to enhance image quality with the intent to emphasise structures in the image, which inform the automated recognition of objects or categories of objects.
Processing to attenuate or discard elements of the image, which are unlikely to be useful for the pattern recognition process.
Processing converts image to a standard format suitable for feature extraction and pattern recognition routines.
Notes – other classification places
Specific aspects of pre-processing are covered by the subgroups of group G06V 10/20; they particularly relate to aspects such as:
- Processes or devices for identifying regions of the image, which should be subjected to the pattern recognition process, or which are likely to contain image information that is relevant for an object recognition task - covered by group G06V 10/22;
- Correcting wrongly oriented images (e.g. changing the orientation from an erroneous portrait mode to landscape mode), compensating for the pose change of the object by performing affine transformations (translation, scaling, homothety, similarity, reflection, rotation, shear mapping, and compositions of them in any combination and sequence), or for correcting geometrical distortions induced by the image capturing - covered by group G06V 10/24;
- Determination of a bounding box containing the pattern of interest, processing within a region-of-interest [ROI] or volume-of-interest [VOI] to emphasise the pattern for recognition – covered by group G06V 10/25;
- Devices or processes for separating a candidate object from other, non-interesting image regions or the background; image segmentation to the extent that it is adapted to support a subsequent recognition step - covered by group G06V 10/26;
- Adjusting the bit depth, e.g. conversion to black-and-white images, and setting thresholds therefor, e.g. by analysis of the histogram of the image grey levels; Converting the image data to a predetermined numerical range, e.g. by scaling pixel values – covered by group G06V 10/28;
- Techniques for improving the signal-to-noise (SNR) ratio or denoising the image for the purpose of improving the recognition – covered by group G06V 10/30;
- Adjusting the size or the resolution of the image to a standard format, e.g. by scaling; adjusting the size of the detected object to a certain format – covered by group G06V 10/32;
- Smoothing or thinning to obtain an alternative, less complex representation of the pattern; applying morphological operators (e.g. morphological dilation, erosion, opening or closing) for filling in gaps or merging elements, with the aim of emphasising the structures relevant for recognition; Skeleton extraction for characterising the shape of a pattern – covered by group G06V 10/34;
- Enhancing the contrast by convolving the image with a filter mask or by applying a non-linear operator to local image patches – covered by group G06V 10/36.
Examples
Alignment of the image of a face by affine transformations to obtain a pose-invariant image
Relationships with other classification places
Different image preprocessing in general are covered in groups as it follows:
- G06T 3/00 when geometric image transformations (e.g. image rotation) are involved;
- G06T 5/00 when image enhancement or restoration (e.g. denoising) is performed.
References
Application-oriented references
Recognising scenes; Scene-specific elements
| G06V 20/00 |
Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
| G06V 30/00 |
Image or video recognition or understanding of human-related, animal-related or biometric patterns in image or video data
| G06V 40/00 |
Informative references
Filter operations to reveal edges, corners, or other image features, which are used to characterise objects
| G06V 10/44 |
Image enhancement or restoration
| G06T 5/00 |
Image segmentation
| G06T 7/10 |
Morphological operators for image segmentation
| G06T 7/155 |
Glossary
DCT
| discrete cosine transform
|
FFT
| fast Fourier transform
|
FOV
| field of view, the region of the environment that an image sensor observes.
|
ROI
| region of interest, an image patch that is likely to contain relevant information.
|
skeletonisation
| process of shrinking a shape to a connected sequence of lines, which are equidistant to the boundaries of the shape.
|
SNR
| signal-to-noise ratio
|
VOI
| volume of interest, a cuboid that encloses three-dimensional data points that are likely to represent relevant information.
|