Abstract: Aspects of the present invention are related to systems and methods for detecting and locating skipped frames in a video sequence.

Abstract: Aspects of the present invention are related to methods and systems for image processing, in particular, to methods and systems for computationally efficient and robust image alignment. Initially, larger image features may be used to approximate an alignment measurement. Results of the initial measurement may be used to narrow the search range used in conjunction with higher-resolution images in subsequent alignment measurements. The narrowed search range may insure that if the alignment measurement using the fine-detail images fails, then the measured results may have the accuracy as indicated by the prior, reduced-image-based measurement.

Abstract: A method of measuring instantaneous signal-dependent nonlinear distortion in a system under test in response to a test signal having a varying frequency sinusoidal component, such as a swept frequency signal or a multi-burst signal, uses a time window to locate a frequency of interest in the corresponding varying frequency component of a signal output from the system under test. Once the location of the frequency of interest is located in the output signal, a spectrum of the portion of the output signal within the time window is generated and the magnitudes of the spectral peaks, including the spectral peak for the frequency of interest or nominal frequency and all isolated peaks representing distortion frequencies, are measured. The amount of distortion is calculated as a ratio of the square-root of the sum of the magnitudes of the distortion frequencies to the magnitude of the nominal frequency.

Abstract: Embodiments of the invention include systems and methods for predicting video quality through human vision perceptual color response and associated perceptual differences of video relative to a reference. Embodiments include a model of the human vision system developed as an extension to CIECAM02 to predict the macro-behavioral response to light stimuli with the following varying parameters, in addition to the parameters relevant to CIECAM02: spatial frequency, temporal frequency, angular extent, temporal extent, and surround.

Abstract: Embodiments of the invention include systems and methods for automatically detecting persons and/or other objects in a video signal. Automatic color compensation is applied to the video signal based on measured color of a reference area of the video signal. Generally the reference area is an area of the video signal that does not change often over time. The color compensated video signal is then analyzed for the appearance of objects and persons. In some instances the persons are identified using facial or other type of recognition. Counters or other trackers may provide real-time information, such as how many people are currently on a particular floor of a building.

Abstract: A variable passband ARMA filter has as inputs a signal to be filtered, a reverse version of the signal to be filtered and a variable coefficient to produce as an output a filtered signal. The variable coefficient is generated by a variable equivalent sample rate coefficient converter having as inputs an initial coefficient and a variable resampling rate parameter which controls the passband.

Abstract: A no-reference subjective quality ratings predictor for a lossy compressed signal decodes the lossy compressed signal to produce a decompressed signal, and extracts from the lossy compressed signal error bounding parameters and information data. An error estimation generator converts the error bounding parameters to sensitivity test data which is combined with lossy data from an inverse compression module within the decoder to produce data with bounded errors. The data with bounded errors is converted into a sensitivity decompressed signal. The decompressed and sensitivity decompressed signals are processed by a full-reference subjective quality rating predictor to produce the subjective quality ratings for the lossy compressed signal. The information data and decompressed signal may also be input to the error estimation generator to generate the sensitivity test data in conjunction with the error bounding parameters.

Abstract: A method of generating an MPEG sequence for multiple component analog video measurements modifies a general purpose encoder and provides an MPEG encoded signal to test a device under test (DUT) component analog output that is valid in both RGB and YPbPr in both pulse and bar and that solves digital RGB cross-talk due to aliasing of Cb and Cr in 4:0:0 and 4:4:4 formats. The general purpose encoder is modified by removing vertical pre-filtering and post-filtering, altering quantization tables alternate scanning for vertical redundancy, and turning off motion estimation. The input to the MPEG encoder is pre-aliased to provide minimum digital cross-talk after MPEG compression by repeating every other sample.

Abstract: A method is provided to predict the location of attention focus probability trajectories due to distractions in a test video. A peripheral sensitivity probability map is created base upon a nominal measurement map, and an attention probability map, which are both based on a test video input. A focus of attention probability map with distraction is produced based upon the peripheral sensitivity map, the nominal measurement map, the attention probability map and a distractibility control input.

Abstract: Picture quality measurement systems and methods are provided for measuring DC blockiness within video blocks. Block boundaries are located within a test video frame. The relative AC differences within each block are measured using a reference video frame, a statistically estimated reference or a default value of the white video level divided by two. An objective DC blockiness map, a subjective DC blockiness map or both may be generated.

Abstract: Embodiments of the present invention relate to a method of measuring a test video frame. A test video input is provided, along with an artifact measurement control, a gradient change measurement is performed based upon the test video input and a gradient change measurement map is provided.

Abstract: A robust video temporal method for registration between test and reference video sequences without a priori knowledge of the respective signal parameters initially produces frame and subimage distillation measurements from the test and reference video sequences. The frame distillation measurements are linearly aligned using a local Pearson's cross-correlation coefficient (LPCCC) image to obtain a best alignment line, each pixel of which represents an LPCCC cross-correlation coefficient between frames of the test and reference video sequences. For each pixel of the best alignment line that is below a threshold, a vertical search is performed in the LPCCC image for a higher cross-correlation coefficient as the best cross-correlation coefficient to achieve temporal mapping between frames of the test and reference video sequences.

Abstract: K2T measurement of video signals, regardless of video format, is determined by generating a graticule using an algorithm having a single variable that is a function of the video format for the particular video signal. The graticule is set to a normalized value of one for a range of +/?T, where T is a function of the video format. The graticule is set to a normalized value over a range of +/?(1.5T to flatBreak) according to a human vision masking model algorithm expressed as K+C/t, where K is a constant K factor, t is a function of T, and C is the single variable that is a function of the video format. The normalized value of the graticule beyond a point designated by flatBreak is a constant=K. Finally a linear curve fit between one and K+C/t over the range of +/?(T to 1.5T) sets the remaining values for the graticule.

Abstract: An up-sampled half band reconstruction filter has a seagull architecture in the form of a pair of parallel infinite impulse response (IIR) filters, one receiving as an input an input signal delayed by one sample time and the other receiving as an input a reverse version of the input signal delayed by one sample time. The outputs from the pair of parallel IIR filters are combined with the input signal to produce a reconstruction filter output in response to the input signal.

Abstract: A method and apparatus are provided to measure spatial distortion. Measure and remove borders; and rescale the active test image, the active reference image, or both, such that they are the same size. A representative line is obtained, or selected, from each image providing a reference representative line (ReferenceRL) and a test representative line (TestRL). A local Pearson' cross-correlation coefficient (LPCCC) image is created from both the ReferenceRL and the TestRL at different horizontal shifts. The highest average intensity line across the LPCCC image is found as the coordinates of the corresponding Hough Transform image pixel with maximum intensity is determined. The Hough Transform Theta and R are converted to spatial scale and offset. The crude and refined offsets and scale values are combined to produce total horizontal and vertical scale offset values. These spatial measurements can then be provided.

Abstract: A method of measuring relative channel delay between components of a video signal includes removing a local mean from each of the pair of input component signals and cross-correlating the resulting pair of input component signals. A mean is removed from the cross-correlation and a centroid for the cross-correlation is found within a region bound by nearest zero-crossing to a peak in the cross-correlation. The centroid is then converted to a delay time as a function of sample rate after removing a sample offset. Also a normalized peak value of the cross-correlation provides a figure of merit for the probable accuracy of the delay measurement.

Abstract: A picture quality diagnostics apparatus and method generates a human vision model response based on a human vision model for a test input video signal. Also objective measure maps for different impairment types are generated from the test input video signal. The objective measure maps are applied as masks to the human vision model response to produce objectively filtered subjective maps. The objectively filtered subjective maps are analyzed to give the respective proportions of the different objective impairment types contributing to perceptual impairment or difference for the test input video signal.

Abstract: A method of frequency response measurement for a sinusoidal test signal, such as a swept sinusoid signal, a multi-burst sinusoidal signal or the like, uses a complex sinusoid window at a particular frequency for correlation with the sinusoidal test signal. The resulting complex correlation magnitude signal is thresholded as a function of a percentage of a maximum complex correlation magnitude. A centroid of the thresholded complex correlation magnitude signal is found, and the complex correlation magnitude at the centroid is the frequency response at the particular frequency.

Abstract: A streaming media quality analyzer system compares original streaming media from a source, or original packetized streaming media from the source, with a reconstructed packetized streaming media representing the packetized streaming media as received over a network in a connectionless manner at a remote site. At the remote site the received packetized streaming media is analyzed to determine time of arrival of the packets including identification of missing packets. The analysis results are transmitted over the network to a measurement site and used to reconstruct the received packetized streaming media from the original packetized streaming media. The reconstructed packetized streaming media is then analyzed to determine the quality of the packetized streaming media received at the remote site.

Abstract: A high-speed digital-to-analog converter (DAC) measurement method acquires and quantizes an analog ramp output by the DAC corresponding to a digital ramp input to produce a quantized ramp, determines a start and end of the quantized ramp, obtains a difference between the quantized ramp and an ideal ramp to produce a quantized periodic signal (triangular or sinusoidal), determines a frequency for a qualified peak from an FFT of the quantized periodic signal, produces a mask filtered periodic signal from an iFFT around the qualified peak, and determines a sample window spanning a local maximum and minimum for each period of the quantized periodic signal. The ramp step levels are the averages of the samples within each sample window. From the step levels in DAC LSB's, resolution, monotonicity, differential lineary and integral linearity are determined for the DAC.