Quality Control Optimizing Automation with Efficient Content Verification Vivek Koul, Project Manager, Interra Systems

The transition to file-based workflows enables automation at each stage in the content lifecycle and transformation of content specific to each stage. With automation comes the challenge of ensuring content quality and hence the need for automated content verification systems or QC (quality control) solutions. QC solutions enhance efficiency and the cost effectiveness of automation.

Process Automation is a paradigm in today’s broadcast industry. The transition to file-based workflows and the advent of new technologies have made automation possible at each stage in the content lifecycle - ingest, transfer, transcoding, and delivery. The challenge lies in automating processes to achieve efficient and cost-effective workflows.

Complexities in Workflows

With the technology advances in capture, editing, storage, and delivery, large amount of media content, with vastly diverse characteristics, is being supplied to the broadcast and post-production facilities. In the workflow, media content is modified by several systems before reaching the end-user. An example of this workflow would be – an ingest system acquires new content, automatically transfers and schedules this content for re-purposing, transcodes the content, and transfers it for final delivery.

An essential ingredient for the efficient operation of such a workflow is the interoperability and synchronization between the different systems and technologies that are operating within it. However, for truly optimum efficiency, content quality and correctness need also to be ensured throughout the workflow.

Complexities in Content Verification

As content moves in the workflow across several technologies and stages, content is transformed – from one format to another, from one resolution to another. Each transformation in the workflow modifies the characteristics of the content and leads to the possible introduction of specific quality issues.

The integrity and readiness of the content needs to be maintained as it is transformed across each workflow. However, content verification is complex. At each stage in the workflow, content verification requirements tend to be different. At a certain stage, basic checks against certain pre- defined specifications may be enough. Then, at another stage, detailed format checking with an in-depth study of quality aspects of the content may be required.

For example:

At Ingest, besides verifying video features such as resolution and color information, capture related issues such as tape-induced block errors, blurriness in motion, background audio noise may also need to be detected.

At Post Production, issues such as cadence break, field dominance errors, and audio level mismatch may need to be detected.

After Transcoding, an in-depth verification of content is required. In addition to verification of basic video features, the syntax of the content must also be validated against the corresponding format specification. Additionally, detection of issues such as blockiness and pixelation becomes critical.

At File Transfer, it may be essential to verify the size and integrity of bits and bytes within the content file.

At the time of Delivery, conformance to delivery specifications such as TR101290, CableLabs, ATSC, DVB, ISMA needs to be verified.

Automating Content Verification

With transformation of content at each stage in its lifecycle, there is a need for a broad-based content verification system. This system needs to be robust, configurable, and easy to fit in with other systems in the workflow. At the same time, the system needs to be scalable to keep pace with the ever-increasing demands for content-processing and performance.

Figure 1: Quality issues with transformation of content 36

Traditionally, content verification has always been viewed as an important operation in the workflow. However, content verification used to be, mostly, a manual operation where verification was done visually. Experts with extensive experience in the intricacies of the quality aspects of video and audio inspected the content and certified it. This method, though effective, was not very efficient. Hours of manual inspection were required to verify and qualify the large amounts of content. Moreover, this verification was mostly done just before play-out or final delivery. A manual verification at every stage is practically impossible and simply not cost effective. An automated system, with built-in intelligence, could prove more efficient, cost-effective, and easy to plug in at various points in the workflow.

AUTOMATION FOCUS