PvaOperatorTypes#

Macros#

PVA_BRIEF_DESCRIPTOR_ARRAY_LENGTH: Brute Force Matcher operator's length of Brief Descriptor Array.
PVA_MAX_BF_DESCRIPTOR_COUNT: Brute Force Matcher operator's maximum number of descriptors for either queries or references.
PVA_MAX_BF_MATCHES_PER_DESCRIPTOR: Brute Force Matcher operator's maximum matches per descriptor.
PVA_RADAR_DOPPLER_FOLD_COUNT: Number of DDM Doppler folds.
PVA_RADAR_MAX_TARGET_COUNT: Maximum number of targets.
PVA_RADAR_NUM_ANTENNA_ELEMENTS: Number of upper or lower antenna elements in the radar array aperture.
PVA_RADAR_NUM_TARGET_DETECTION_PROPERTIES: Number of target detection properties (7 values: velocity, range, azimuth, elevation, x, y, and z)
PVA_RADAR_NUM_TOTAL_ANTENNA_ELEMENTS: Number of total antenna elements.
PVA_RADAR_PEAKDET_NUM_PEAK_INDICES: Peak Detection operator's number of peak indices (3 values: range bin index, folded doppler bin index, and unfolded doppler bin index)
PVA_RADAR_RX_ANTENNA_COUNT: Number of receive antennas (4T4R radar configuration)
PVA_RADAR_TX_ANTENNA_COUNT: Number of transmit antennas (4T4R radar configuration)

Enumerations#

PVABartlettAzNeighborhood: Azimuth neighborhood size for Separable Bartlett beamforming.
PVABartlettAzSource: Azimuth source selection for Separable Bartlett beamforming.
PVABatchFFTWindowType: Batch FFT window types.
PVAConnectivityType: CCL Connectivity Types.
PVADistanceType: Defines the types of distance metrics for the DistanceTransform operator.
PVADoaAngleFFTPrecision: Angle estimation precision levels for DOA Angle FFT operator.
PVADopplerFFTOutputLayout: Output layout options for Doppler FFT operation.
PVAFlipDirection: Flip direction options for image flip operation.
PVAGMMDataType: Data types for the Gaussian Mixture Model (GMM) used in background subtraction.
PVAImageStatFlag: Image Statistics operation flags.
PVAInterpolationType: Interpolation method types for image operations.
PVAMixChannelsCode: MixChannels Operation Codes.
PVAMorphologyType: Morphology Operation Codes.
PVARadarCFARType: Radar CFAR algorithm types for target detection.
PVAWarpTransformationType: Warp perspective operator's transformation types.
PvaRadarCFARHorizontalThresholdType: Configuration parameters for radar CFAR target detection operator.

Structs#

PVABFMatchesTypeRec: Structure containing match information for a single query descriptor.
PVABriefDescriptor: Brute Force Matcher operator's structure containing the BRIEF Descriptor.
PVADoaAngleFFTParams: Parameters for the DOA Angle FFT and Target Processing operator.
PVADoaBartlettBeamformingParams: Parameters for the Bartlett Beamforming DOA + Target Processing operator.
PVADopplerFFTParamsRec: Doppler FFT parameters.
PVAImageStatOutput: Image Statistics output structure.
PVARadarGP: Radar processing parameters structure for DOA (Direction of Arrival) target processing.
PVARangeFFTParamsRec: Range FFT parameters.
PVASeparableBartlettParams: Parameters for Separable Bartlett beamforming (sequential 1D DOA estimation).
PVASnapshotExtractionParams: Parameters for the Snapshot Extraction operator.
PVATargetProcessingParams: Parameters for the Target Processing operator.
PvaRadarCFARHorizontalThresholdRec: Horizontal CFAR threshold specification.
PvaRadarCFARParamsRec: Parameters for the PVA Radar CFAR (Constant False Alarm Rate) operator.

Enumerations#

enum PVABartlettAzNeighborhood#

Azimuth neighborhood size for Separable Bartlett beamforming.

Controls how many azimuth bins around the Step 1 peak are searched in Step 2. At boundaries the window is clamped to valid indices.

Values:

enumerator PVA_BARTLETT_AZ_NEIGHBORHOOD_0#: Only peak azimuth bin, complexity O(Naz + Nel)

enumerator PVA_BARTLETT_AZ_NEIGHBORHOOD_1#: ±1 bins (3 azimuth bins), complexity O(Naz + 3*Nel)

enumerator PVA_BARTLETT_AZ_NEIGHBORHOOD_2#: ±2 bins (5 azimuth bins), complexity O(Naz + 5*Nel)

enum PVABartlettAzSource#

Azimuth source selection for Separable Bartlett beamforming.

Controls where the final azimuth estimate comes from in the two-step search.

Values:

enumerator PVA_BARTLETT_AZ_FROM_NEIGHBORHOOD#: Final azimuth and elevation both come from Step 2 (elevation scan). The joint peak of the local (az neighborhood × elevation) patch is used.

enumerator PVA_BARTLETT_AZ_FROM_EL0#: Azimuth comes from Step 1 (azimuth scan at el ≈ 0°); only elevation is taken from Step 2. Can give better azimuth accuracy when targets are near the horizon; may degrade at steep elevation angles.

enum PVABatchFFTWindowType#

Batch FFT window types.

Values:

enumerator PVA_BATCH_FFT_WINDOW_HANNING#: Hanning window.

enumerator PVA_BATCH_FFT_WINDOW_USER_DEFINED#: User defined window.

enum PVAConnectivityType#

CCL Connectivity Types.

Values:

enumerator PVA_4_CONNECTIVITY#: 4-connectivity

enumerator PVA_8_CONNECTIVITY#: 8-connectivity

enum PVADistanceType#

Defines the types of distance metrics for the DistanceTransform operator.

Specifies the different distance calculation methods available.

Values:

enumerator PVA_DIST_L1#: Manhattan distance (L1 norm).

enumerator PVA_DIST_L2#: Euclidean distance (L2 norm).

enumerator PVA_DIST_HAMMING#: Hamming distance.

enum PVADoaAngleFFTPrecision#

Angle estimation precision levels for DOA Angle FFT operator.

This enum controls the zero-padding applied to the virtual antenna array snapshot before performing the spatial FFT for angle estimation. The precision level determines the FFT scaling factor (zero-padding factor), which affects the density of the frequency grid in the FFT output.

Zero-padding produces a denser frequency grid, which is equivalent to sinc interpolation in the angle domain. This enables more precise localization of spectral peaks without increasing the fundamental angular resolution (which is determined solely by the physical antenna aperture).

The FFT lengths are computed as:

Azimuth FFT length: numVirtualAzimuthElements * scalingFactor
Elevation FFT length: numVirtualElevationElements * scalingFactor

Values:

enumerator PVA_ANGLE_ESTIMATION_PRECISION_LOW#: Minimal zero-padding, fastest.

enumerator PVA_ANGLE_ESTIMATION_PRECISION_DEFAULT#: Balanced precision and performance.

enumerator PVA_ANGLE_ESTIMATION_PRECISION_HIGH#: Maximum precision, slowest.

enum PVADopplerFFTOutputLayout#

Output layout options for Doppler FFT operation.

Values:

enumerator PVA_DOPPLER_FFT_OUTPUT_LAYOUT_DOPPLER_RX_RANGE#: Output layout: (Doppler bins x Rx channels x range bins)

enumerator PVA_DOPPLER_FFT_OUTPUT_LAYOUT_RANGE_RX_DOPPLER#: Output layout: (range bins x Rx channels x Doppler bins) - transposed.

enumerator PVA_DOPPLER_FFT_OUTPUT_LAYOUT_RANGE_DOPPLER_RX#: Output layout: (range bins x Doppler bins x Rx channels) - transposed with L2 buffering.

enum PVAFlipDirection#

Flip direction options for image flip operation.

Values:

enumerator PVA_FLIP_HORIZONTAL#: Flip image horizontally (left-right)

enumerator PVA_FLIP_VERTICAL#: Flip image vertically (up-down)

enumerator PVA_FLIP_BOTH#: Flip image both horizontally and vertically (180° rotation)

enum PVAGMMDataType#

Data types for the Gaussian Mixture Model (GMM) used in background subtraction.

Specifies the available formats for storing GMM parameters such as weight, variance, and mean.

Values:

enumerator PVA_GMM_DATA_TYPE_FP32#

enumerator PVA_GMM_DATA_TYPE_FP16#

enum PVAImageStatFlag#

Image Statistics operation flags.

Values:

enumerator PVA_IMAGE_STAT_FLAG_PIXEL_COUNT#: Calculate pixel count.

enumerator PVA_IMAGE_STAT_FLAG_SUM#: Calculate per-channel sum.

enumerator PVA_IMAGE_STAT_FLAG_MEAN#: Calculate per-channel mean, triggers PVA_IMAGE_STAT_FLAG_SUM and PVA_IMAGE_STAT_FLAG_PIXEL_COUNT.

enumerator PVA_IMAGE_STAT_FLAG_VARIANCE#: Calculate per-channel variance, triggers PVA_IMAGE_STAT_FLAG_MEAN.

enumerator PVA_IMAGE_STAT_FLAG_COVARIANCE#: Calculate full covariance matrix, triggers PVA_IMAGE_STAT_FLAG_VARIANCE.

enum PVAInterpolationType#

Interpolation method types for image operations.

Values:

enumerator PVA_INTERPOLATION_NN#: Nearest neighbor.

enumerator PVA_INTERPOLATION_LINEAR#: Bilinear interpolation.

enum PVAMixChannelsCode#

MixChannels Operation Codes.

Mix channels is a copy operation from a set of input channels to a set of output channels.

The set of inputs and outputs may be given by any number of input and output images, where each image may have one or more channels.

The channel mapping from input to output is given by two arrays, one with the indices of the input channels, and the another with the corresponding channel index in the output.

The channel indices are enumerated starting from zero and increasing monotonically across all channels of all images provided in the array. For example, given an array of 3 RGB images, index 5 corresponds to the B channel of the second image.

Values:

enumerator PVA_SPLIT_RGBA8_TO_U8#: split 1 rgba8 image into 4 u8 images, [0, 1, 2, 3] -> [0, 1, 2, 3]

enumerator PVA_MERGE_U8_TO_RGBA8#: merge 4 u8 images into 1 rgba8 image, [0, 1, 2, 3] -> [0, 1, 2, 3]

enum PVAMorphologyType#

Morphology Operation Codes.

Values:

enumerator PVA_ERODE#: Erode operation.

enumerator PVA_DILATE#: Dilate operation.

enum PVARadarCFARType#

Radar CFAR algorithm types for target detection.

This enumeration defines the available Constant False Alarm Rate (CFAR) algorithms that can be used for radar target detection. Each algorithm provides different approaches to estimate the noise floor and set adaptive thresholds for target detection.

Values:

enumerator PVA_CFAR_CA#: Cell Averaging (CA) algorithm - Thresholds are set by averaging the signal values from both leading and trailing training cells.

enumerator PVA_CFAR_CA_OS#: Order Statistic (OS) algorithm - Thresholds are set using the median value of the ranked signal amplitudes from both leading and trailing training cells.

enumerator PVA_CFAR_CA_GO#: Greatest Of (GO) algorithm - Thresholds are set using the greater of the averages from the leading and trailing training cells.

enumerator PVA_CFAR_CA_SO#: Smallest Of (SO) algorithm - Thresholds are set using the smaller of the averages from the leading and trailing training cells.

enum PVAWarpTransformationType#

Warp perspective operator’s transformation types.

Values:

enumerator PVA_WARP_TRANSFORMATION_PERSPECTIVE#: Perspective transformation.

enumerator PVA_WARP_TRANSFORMATION_AFFINE#: Affine transformation.

enum PvaRadarCFARHorizontalThresholdType#

Configuration parameters for radar CFAR target detection operator.

This structure contains all the necessary parameters to configure the Constant False Alarm Rate (CFAR) algorithm for radar target detection. The parameters control the detection sensitivity, algorithm type, and processing window dimensions for optimal target detection performance.

The CFAR algorithm uses training cells to estimate the noise floor and guard cells to protect the cell under test (CUT) from target energy leakage, ensuring accurate threshold calculation.

CFAR Cell Arrangement: N = numTrain K = numGuard

* | T1 | T2 | T3 |......| TN | G1 | ... | GK | CUT | G1 | ... | GK | T1 | T2 | T3 |......| TN |
* |<-------- Trailing ------>|<--- Guard --->|     |<--- Guard --->|<-------- Leading ------->|
*

Selects the mode for the horizontal CFAR threshold factor.

Passed as the kind field of PvaRadarCFARHorizontalThreshold.

Values:

enumerator PVA_CFAR_HORIZONTAL_THRESHOLD_UNIFORM#: Single factor applied uniformly to all range rows.

enumerator PVA_CFAR_HORIZONTAL_THRESHOLD_PER_ROW#: One factor per range row (range-adaptive)