How to use image quality analyzer to measure system performance of high-definition images

　　The purpose of HD (High Definition) TV systems is to create a new quality of experience (QoE). This QoE is determined by several factors, the most important of which are image and sound quality. There is an inverse relationship between quality and content. When the quality is the same, the content becomes the key to determine the quality of the program, and when the content is the same, the quality becomes the determining factor. In any case, TV is always a business behavior, and each operator must determine a strategy to balance the contradiction between content and quality.

　　The purpose of testing and measurement is nothing more than to improve and manage program quality. Similar to the automotive industry, we must accept the fact that quality control is not a process that can be tinkered with after the program is produced, but must be started from the beginning and implemented at every step of the program production process.

　　While the quality of analog signals and full-bandwidth digital video can be characterized indirectly by measuring the distortion of static test signals, compressed TV systems present a more daunting challenge. Picture quality in a compressed system can vary dynamically with the data rate, image complexity, and encoding algorithm used. The static nature of the test signal does not reflect the true characteristics of picture quality. Test scenes with normal content and motion can be used to measure signal quality and the results can be determined based on viewer reports, but this is a woefully inefficient method of evaluating the performance of compressed video systems. The Tektronix Picture Quality Analyzer PQA500 provides a fast, practical, repeatable, and objective measurement solution for subjective assessment of picture quality.

　　PQA500 Introduction

　　The input of PQA500 is two video files: an original reference video sequence file and a compressed or processed damaged reference file. First, PQA500 can align the two input sequences in space and time without the calibration stripes embedded in the video sequence. Then, the quality of the test video is analyzed according to the human visual system and attention model, and a quality measurement result that is highly correlated with subjective evaluation is output, including an overall quality report of the input video, frame-by-frame measurement results, and a damage map of each frame. In addition, PQA500 also provides traditional image quality measurement data such as PSNR (peak signal-to-noise ratio) as an industry benchmark, as well as an image damage diagnosis toolkit for measuring typical video damage and detecting image defects.

　　Each reference video sequence and test clip has a different resolution and frame rate. PQA500 can measure image quality between HD and SD, between SD and CIF, and in many other combinations. PQA500 also supports measurement clips of unlimited duration, allowing quantitative testing of image quality over a full movie through various conversion techniques.

　　SMPTE 372M Standard

　　To obtain the best quality video material, 1080p50 or 1080p60 formats are the best choices. Since these two formats require RGB 4:4:4 processing in post-production, a high-speed interface is needed to handle the additional data. Currently, most equipment relies on dual link to implement this high-speed interface, but the SMPTE 372M dual link standard supports multiple modes and color spaces when transmitting video and audio data. To provide first-class measurement results, single channel analysis, channel-to-channel timing measurements, and combined waveforms and vectors for color grading applications are required. Tektronix's WFM/WVR7120 modular waveform monitoring product series can provide detailed analysis for such applications, with enough accuracy to make reliable physical layer measurements at 1.5Gbit/s.

　　Lip Sync Problems

　　The most annoying impairment in a TV signal is the desynchronization of the audio and video. This problem is known in the industry as lip sync. It occurs when the timing of the audio processing differs from the video processing and the two are not properly aligned.

　　There have been attempts to address this problem using online testing and correction methods, but most encoding schemes actually cut off the ancillary data field, where the traditional online test signal is located. This problem has become more difficult to solve with the increasing popularity of multi-channel audio and surround sound audio in compressed formats.

　　Therefore, offline testing using a set of specific audio and video programs is still an option. Although most signals use test patterns that rely on the operator's hearing and vision to find the correct timing relationship between audio and video, the award-winning test programs provided by Tektronix are now not only available in high-end analyzers, but are also integrated in the new WFM7120.

　　When using this test procedure, a set of related audio and video test signals are first sent into the system, and then the waveform monitor accurately measures the timing difference between the audio and video. The test signal supports SD and HD digital formats.

　　Cross-layer monitoring

　　The key to the commercial success of HDTV lies in whether it can provide the best image quality for viewers. The leading standard in satellite and cable TV applications is H.264/MPEG-4 Part 10, while VC-1 and H.264 dominate the IP field.

　　To fully exploit all the potential benefits of these video codecs, the signal transmission channel must be properly tuned, whether it is satellite, cable, terrestrial or IP transmission. In mobile TV, lower resolution formats such as CIF or QCIF are used, but the problems are very similar to those of IP channels. For traditional satellite, cable or terrestrial broadcast systems, attention must be paid to both the RF domain and the video domain.

　　This approach, called cross-layer monitoring, is a pre-processing method that allows for rapid fault isolation and root cause identification when transmission problems are encountered. To achieve economies of scale, Tektronix has introduced RF channel rotation in its instruments, which allows for intelligent, cyclic detection between channels while still giving priority to channels of higher interest or value. In the IP domain, rapid monitoring across many segments is now possible, which can immediately indicate important signal failures and provide in-depth diagnostics, helping designers find the right problem.

　　The MTM400 MPEG, RF and IP monitors ensure that quality of service requirements and service level agreements are tracked and met.

　　Basic data flow analysis

　　In order to improve H.264 quality, Tektronix has also developed a test suite for basic data stream analysis to improve programming efficiency. The suite includes an H.264 multiplexer, a buffer analyzer and a transport stream analyzer. The H.264 multiplexer can help users create test transport streams using Tektronix's Vclips test data stream library, the buffer analyzer is used to verify the consistency of set-top box design and data streams, and the transport stream analyzer is used to ensure that the data streams encoded with H.264 are transmitted.

　　The test suite can be used with test hardware such as MT430 for real-time measurement. In order to improve connectivity and cross-layer testing capabilities, the upgraded MT430 has multiple RF and IP interfaces.

　　Supporting end-to-end testing of HDTV requires a variety of test tools and expertise in different domains, such as uncompressed video and audio in multiple formats, compressed video for playback, and file-based video for storage and workflow.