Defining-Video-Quality-Requirements-508.pdf

Defining Video Quality Requirements A Guide for Public Safety

May 2013

Initial Release: July 2010

Defining Video Quality Requirements:

A Guide for Public Safety

Volume 1.0

Developed by:

Video Quality in Public Safety Working Group

This procurement guide was sponsored by the Office for Interoperability and Compatibility within the U.S.

Department of Homeland Security. It was developed in partnership with emergency responders, the U.S.

Department of Commerce’s Public Safety Communications Research program, and private sector partners.

Defining Video Quality Requirements: A Guide for Public Safety Volume 1.0

Executive Summary Often, emergency responders must consider a multitude of factors, such as installation, testing, support, redundancy, and training, before making video component procurement decisions. The guide provides an overview of video systems, defines functional concepts of video quality, explains how to generalize a use case with a use class, and provides a brief explanation of the qualitative aspects of video components. Emergency responders involved in the procurement process—of a video system either in part or in its entirety— will find this guide valuable because it considers an end---to---end system. For example, this guide identifies needs associated with a video stream as it travels from the scene (camera) through the system to the end user viewing the scene on a remote display.

There is a wide range of information that exists related to video quality and selection of video components such as installation, maintenance, training, and interoperability. While this guide does not provide detailed specifications and standards for video components, the Video Quality in Public Safety Working Group plans to release future guidance on technical performance specifications and standards that address various components of the video system.

Use Cases

A video system must deliver video to the end users in such a way that they are able to accurately recognize objects and take action based on what they see. The first step in procuring video equipment is to determine the use case. A use case is a set of functional requirements based on the content of the observed scene and the task being performed by an end user. For instance, the scene of interest could be a large crowd on a train platform, a license plate on a moving vehicle, or a plume of smoke, and an end user’s task might be to view a scene in real time through the video apparatus and alert emergency responder officials if something is awry. Alternatively, the video apparatus may capture and record footage for an end user to review at a later date. These applications for use are called “use cases.” This guide will introduce users to different use cases, enabling users to determine which level of quality is necessary and making it possible to procure an appropriate video system with the required level of video quality.

Use Classes

Although a particular video use case may seem unique, it almost certainly has features common to other use cases. Use classes represent combinations of shared features across various use cases. Video footage from a police car and aerial video footage of a wildfire, although different in many respects, are similar in that they are capturing an image that is moving at a high degree. By answering five questions about a particular use case, the reader can determine which general use class applies. Key aspects to consider include

1. Usage Time Frame – Is the video used for real---time applications or recorded for later use?

2. Discrimination Level – What is the end user’s ultimate goal?

3. Target Size – How much of the frame does the object or person of interest occupy?

4. Motion – How much motion (either target or camera) and how much spatial detail are in the video frame?

5. Lighting Level – Is the lighting generally uniform or are there near---black to daylight ranges in the video frame?

ii

Core Video System Components

In addition to understanding use cases and use classes in the procurement of video systems, it is also important to understand the various components of video systems and which components impact quality. To address those issues, this document describes various core video system components that directly affect video quality, such as: Lens Configuration, Image Capture, Processing, Transport, Storage, and Display.

Qualitative Guidance

Lastly, the document provides qualitative guidance to help the reader understand features of system components and define the impact those components have on video quality based on an identified use class. This guidance is intended to educate readers on the issues involved, tradeoffs, and units of measure associated with video components. This information informs the reader how certain use cases and scene content can affect video component features.

Table of Contents Defining Video Quality Requirements: A Guide for Public Safety .......................................................................... i Executive Summary ................................................................................................................................................. ii Table of Contents ................................................................................................................................................... iv List of Figures ........................................................................................................................................................... v List of Tables ............................................................................................................................................................ v I. Introduction

A. Background B. Problem Statement C. Intended Audience D. Document Scope E. How to Use This Guide

II. Fundamental Concepts A. Video Quality B. Use Case C. Generalized Use Class

III. Generalize a Use Case into a Use Class A. Generalized Use Class Aspects B. Generalized Use Class Questionnaire C. Generalized Use Class Examples

IV. Core Video System Components A. Lens Configuration B. Image Capture C. Processing D. Transport E. Storage F. Display

V. Qualitative Guidance A. Usage Timeframe B. Discrimination Level C. Target Size D. Motion E. Lighting Level

VI. Ongoing VQiPS Efforts VII. Conclusion Appendix A: Glossary Appendix B: Acronyms Appendix C: Acknowledgements

List of Figures Figure 1. How to Get the Appropriate Video Quality Figure 2. Use Case Figure 3. Multiple Use Cases Have Aspects in Common Figure 4. Example of Target Size: Large Figure 5. Example of Target Size: Small

Figure 6. Example of Motion in the Scene: High Figure 7. Example of Lighting Level: Variable Figure 8. Generalized Use Class Aspects Figure 9. Application 1: Generalized Use Class Figure 10. Application 3: Generalized Use Class Figure 11. Core Video System Components

List of Tables Table 1. Generalized Use Class Questionnaire

I. Introduction In recent years, the exponential growth of Closed Circuit Television (CCTV) and other video technology has been instrumental in addressing a wide range of emergency response concerns. The emergency response community, including police, fire, and emergency medical services (EMS) are using video applications in transportation, crime, and public works efforts. With emergency responders increasingly relying on CCTV technology and systems, they need to increase their attention on video quality issues so that they can use and share images across interoperable emergency response systems.

To respond to this growing need, the Office for Interoperability and Compatibility (OIC) within the U.S.

Department of Homeland Security worked with the U.S. Department of Commerce’s Public Safety Communications Research (PSCR) program to develop a partnership of local, state, and Federal representatives to address these concerns. These representatives partnered with members of the private sector and academia to form the Video Quality in Public Safety Working Group (VQiPS WG), which is working to enhance and standardize video quality for public safety.

The VQiPS WG has developed this guide to inform and assist the emergency response community in identifying which video system components impact video quality. The guide is intended for emergency responders who are involved in procuring and maintaining a video system.

A. Background At the first VQiPS conference in February 2009, OIC and PSCR initiated the VQiPS WG and defined how the group would address the most pressing video quality issues. The VQiPS WG includes local, state, and Federal representatives from a variety of disciplines, including law enforcement, fire, EMS, transportation, and telecommunications. The VQiPS WG also includes representatives from non---profit research organizations, academic institutions, and the video industry.

The activities of the VQiPS WG and the research and development efforts by OIC and PSCR are all part of a collaborative process to provide representatives of local, tribal, state, and Federal governments with information that will assist them in the procurement of video components and systems. OIC partners with PSCR to support research, development, testing, and evaluation of video system standards and performance requirements. PSCR uses a structured approach to identify video quality needs and specifications and conducts research and lab testing to identify performance specifications for the core video system component areas.

With an end goal to articulate a common set of requirements for the procurement of video components and systems that meet the specific needs of end users, the VQiPS WG has two main purposes: 1) to provide the emergency response community with the knowledge they need to purchase and employ appropriate video systems; and 2) to collectively communicate emergency responders’ needs to industry and standards---making bodies. In order to achieve those purposes, the VQiPS WG developed an action plan to complete the following key initiatives:

1. Develop a set of application---independent usage scenarios.

2. Develop guidance materials to help emergency response communities perform the following:

a. Assess video quality needs.

b. Match needs to technical performance specifications and standards to support procurement.

3. Develop a glossary of common terms.

4. Compile an inventory of existing standards and specifications that address various components of the video system for specific usage scenarios.

5. Develop a common library of test clips that represent the use cases.

This document addresses initiatives 1, 2a, and 3 from the action plan. Currently, the VQiPS WG is addressing the remaining elements (2b, 4, and 5) and will make the results available to the emergency response community in the near future.

B. Problem Statement Video applications are quickly emerging as an essential component of effective emergency response communications. For example, emergency responders can use video to provide aerial images of wildfires to firefighters, to monitor highway traffic, to assess the scene of an accident, or as evidence in a criminal case. In the past, practitioners relied on manufacturers to build and provide video equipment and system specifications. However, video technology has evolved, resulting in more complex video systems and a multitude of video systems and components from which to choose in the procurement process. Because many emergency response agencies do not have the tools, support, and information they need to make informed video system purchasing decisions, they have a major challenge in procuring video systems that attain the desired video quality.1

C. Intended Audience This document addresses the needs of emergency responders who intend to procure video systems, regardless of their organization size, system size, or specific video application. There are two specific categories of potential users of this guide:

1. Emergency response practitioners or persons representing a group of end users within an agency involved in evaluating video systems. This guide will aid in determining video quality needs and selecting core video system components.

2. The technical employee or support contractor (not the end user) who is writing specifications for video systems and developing procurement documents. This guide can serve as a communications tool between an end user and integrator.

This guide will assist those individuals who are responsible for evaluating and procuring video systems that meet the needs of end users. Because there are often multiple end users, the minimum quality requirements should be established, when feasible, for the user with the most stringent video quality needs that are consistent with the most critical use of the system.

D. Document Scope This guide will enable emergency response agencies to more clearly describe an agency’s video quality needs. It will provide qualitative guidance for key video system components and provide a better understanding of which video components impact video quality. The document aims to educate the end user on common components of video systems, which will enable them to purchase the appropriate

1 “Video quality” is the ability of the emergency response agency to use the required video to perform the purpose intended. For example, if the purpose of the video is to capture vehicle license plates in a range of outdoor conditions, video quality is measured in the ability of the video outputs to provide video system. Because this guide will not directly advise users on which video system to purchase, readers should consult with experienced video professionals to identify specific systems and obtain information on jurisdictional procurement regulations.

This document addresses the core components of a video system and the aspects of the scene content that apply to any video system, regardless of size. The term “core components” refers to equipment in the video system that can change the video signal, thereby affecting the quality of the delivered video.

Other considerations for a video system include, but are not limited to, camera placement, security, and weatherization, and are not covered by this guide. Highly complex or unique video applications may need additional requirements that are outside the scope of this guide. For example, the addition of a complex package of video analytics for use with the video system may require additional research and consideration in the preparation of procurement specifications.

This guide is a first step in providing the end user with guidance in selecting video components that affect video quality, such as installation, maintenance, training, and interoperability. In future documents that build on this guide, OIC, PSCR, and the VQiPS WG plan to provide detailed guidance on technical performance specifications and standards that can be applied to various usage scenarios.

E. How to Use This Guide Following the introduction of key concepts, this guide will address two things: first, the steps end users can take to generalize a use case; and second, how to capture user needs and use cases to define the Generalized Use Class. This process is illustrated in Figure 1.

Figure 1. How to Select the Appropriate Video Quality

Chapter 2 outlines the fundamental concepts used in the document. Chapter 3 contains a questionnaire designed to help the reader identify needs in the form of a Generalized Use Class, which is referenced in the remaining sections. Chapter 4 describes the core video systems components. Chapter 5 provides qualitative information that relates back to the Generalized Use Class developed in Section 3. Appendix A contains an extensive glossary that defines commonly used technical terms.

II. Fundamental Concepts This section defines video quality, use case, and a Generalized Use Class.

A. Video Quality Video quality is the ability of the emergency response agency to utilize the required video to perform the purpose intended.

Other terms used to refer to video quality are “visual intelligibility” and “visual acuity.” These equate to a visual extension of a “Can you hear me?” audio test, rephrased as “Can you see me?” This guide reflects this concept as it allows a user to specify end---to---end system performance or individual component performance. This guide is based on the following key concepts:

Every application is trying to recognize a desired target to a particular level of discrimination.

The ability to perform a specified task determines the video quality required.

B. Use Case To ensure the video quality meets end users’ needs, the first step is to clearly define functional requirements or the use case(s). To define the use case, answer these two questions:

1. What is in the scene of interest or scene content?

2. What is the desired task to be accomplished from viewing that scene?

All emergency response video systems must present a scene of interest to a remote user in sufficient detail for the user to make a decision or perform a task based on recognition of what is happening in the scene. For example, the end user must be able to read the characters in a license plate or determine the identities of individuals at a local convenience store while performing surveillance.

Figure 2. Use Case

Figure 2 illustrates the role that video equipment plays in bringing scene content back to the end user.

The performance required of the video equipment depends on the specific use case. A use case considers the scene captured by the video system and the task being performed by the emergency response practitioner (or end user). Figure 2 shows a use case in which an agency has installed video cameras at a train station. The scene content comprises the platform, the train (when it is in the station), and the individuals on the platform. Understanding the end user’s needs for video feeds coming into the command center represents a critical component in assessing the type of video system required. For example, what are the feeds principally used for? Are they being used to monitor the arrival of trains (perhaps to ensure tracks are clear and to notify service and response personnel of an incoming influx of persons into the station)? Are they being used to monitor the flow of passengers on the platform (to look for disturbances or guide response assets in the event of an emergency)? Or are they being used to positively identify and conduct surveillance on suspicious individuals in the crowd and follow them through the station? Later sections of this guide will define how users can identify the aspects of the scene content and use those impact requirements to inform which core video equipment components are necessary.

C. Generalized Use Class This guide’s fundamental premise is that use cases for seemingly different applications have similar quality requirements. This means that, upon closer examination, seemingly disparate video applications may actually have the same minimum requirements to perform a desired recognition task. While one video application may seem unique, it almost certainly has features common to other video applications. Figure 3 illustrates this idea. For example, identifying license plates and monitoring wildfires seem to be very different applications; however, they have many common minimum video requirements. Thus, users can develop performance specifications based on a definite set of Generalized Use Classes. This eliminates the need to create separate performance specifications for every possible application because there are specific Generalized Use Classes that exist.

Figure 3. Multiple Use Cases Have Aspects in Common

Using this theory, this guide presents a finite set of questions that will lead the reader to identify an appropriate Generalized Use Class. These questions are related to how the user intends to use the video images and includes questions related to common aspects such as target size, motion, lighting level, usage time frame, and discrimination level. The following section will provide a questionnaire to determine these common aspects.

III. Generalize a Use Case into a Use Class This section provides examples and explains how to translate the reader’s unique use case into a Generalized Use Class. A description of common aspects that make up a Generalized Use Class, such as discrimination level, time frame, target size, motion, and lighting, is followed by a questionnaire designed to guide the reader through a process of choosing values related to these common aspects.

The particular combination of values serves as an easily identifiable Generalized Use Class that describes an agency’s video quality needs.

An agency or organization may have more than one use case associated with its mission. Each use case should be carefully defined and analyzed as a use class in order to determine the overall most restrictive use class. For example, if law enforcement and fire officials share a video system, law enforcement may need to recognize individuals with high---level detail, while fire may only need to recognize the presence of smoke. In this case, law enforcement has a more restrictive use case.

After the reader has used the questionnaire to determine the Generalized Use Class, subsequent sections of the guide refer to the Generalized Use Class to provide qualitative guidance for the core video system components. The VQiPS WG will make more detailed standards and performance specifications available in subsequent publications.

A. Generalized Use Class Aspects A use class consists of important aspects of specific use cases that are common to other use cases, and generalizations can be made between them. As shown in Figure 2, all use cases have an analyst and a scene under observation. The generalized aspects are derived from both the analyst’s intended use of the video (Use Characteristics) and what is in the scene under observation (Scene Content). The next section describes five Generalized Use Class aspects, each of which can impact the quality of the video content as seen and used by the end user.

Use Characteristics

1. Discrimination Level*

Video may be used to identify a wide range of detail, from motion detection to positive identification of a person for forensic evidence. Not every video system needs to perform positive identification. The video system selected should conform to the application requirements, without over--- or under---specifying the system.

2. Usage Time Frame

To what level of discrimination does the user need to recognize the target?

If a user needs to determine only whether there are people present in a scene, “General Elements of the Action” would suffice. If the user requires large---scale recognition, such as the distinction between a car

Term is defined in Appendix A: Glossary and a van, then “Target Class Recognition” fits the user’s system needs. A value of “Target Characteristics” indicates the need to recognize gender and markings and distinguish smaller actions, and “Target Positive ID” indicates the most specific discrimination level.

Scene Content

3. Target Size

The size of the region of interest (target) with respect to the size of the field of view directly affects the ability to recognize that target when the camera is at its maximum optical zoom. The larger the target, relative to the field of view, the more details can be discerned. The perceived target size is dependent on the actual target size, its distance from the camera, and the camera’s field of view. Figures 4 and 5 demonstrate the difference between a large target size and a small target size. PSCR is developing measurements to more precisely define target size.

Figure 4. Example of Target Size: Large Figure 5. Example of Target Size: Small

4. Motion in the Scene (e.g., background, target, or camera)

Motion can come from the target (e.g., a car driving by), the background (e.g., a large crowd), or from the camera itself moving (e.g., a dash---mounted camera in a police car). Motion affects the length of time a desired target is shown in the video frame and can cause the target to blur. High motion can be caused by either many moving objects within the scene (see Figure 6) or a single object moving quickly.

Figure 6. Example of Motion in the Scene: High

5. Lighting Level

Lighting levels can vary from very dark (e.g., nighttime or indoors) to very bright (e.g., daylight or spotlight), affecting the ability of the camera to capture the image. The presence of both very bright areas and very dark areas in the frame simultaneously is known as high dynamic range, and can impair target recognition, as shown in Figure 7.

Figure 7. Example of Lighting Level: High Dynamic Range

The combinations of the end user’s specific needs in these five areas comprise the Generalized Use Class. There are many more aspects that affect video, and each of these parameters has an infinite number of possible values. This document addresses the most important aspects and defines a manageable number of choices for each. VQiPS WG made a concerted effort to provide a balance between complexity and simplification.

Given the aspects and possible values described above, Figure 8 illustrates the aspects that make up a Generalized Use Class.

Figure 8. Generalized Use Class Aspects

The end user determines the specific value for each aspect. The combination of the individual choices form the user’s Generalized Use Class.

B. Generalized Use Class Questionnaire Table 1 contains the Generalized Use Class questionnaire, which identifies each of the aspects with a brief description, followed by a small set of choices for each aspect. The user should match the appropriate video application with the appropriate Generalized Use Class. The user should use the most restrictive case when making selections.

Table 1. Generalized Use Class Questionnaire

Aspect Question for End User Value Definition Example

Use Characteristics

Discrimination Level

What is the end user’s ultimate goal?

o General Elements of the Action o Target Class Recognition o Target Characteristics o Target Positive ID

High---level description of actions that took place

Large---scale recognition

Medium---scale detail recognition Enough detail to make a positive recognition

People or person present

Car vs. van

Gender, markings, smaller actions Face, object, alpha---numeric

Usage Time Frame

Is the video used for real---time applications or recorded for later use?

o Live or Real---time o Recorded

The video will be viewed at the same time it is being shot

The video will be saved and capable of being played back

Security monitor being viewed by security personnel

Forensic video

Scene Content

Target Size

How much of the frame does the object or person of interest occupy?

o Large o Small

The target occupies a large percentage of the frame

The target occupies a smaller percentage of the frame

Motion

How much motion (either target or camera) and how much spatial detail are in the video frame? If the camera is required to pan---tilt---zoom, the scene complexity could be higher.

o High o Low

There is a lot of motion or edges in the video

Low complexity; there is not much motion or many edges

Lighting Level

Is the lighting generally uniform, or are there near---black to daylight ranges in the video frame?

o Constant Lighting – High o Constant Lighting – Low o Variable

At a comparatively bright level

At a comparatively dim level

Ranges from bright to dim

C. Generalized Use Class Examples The following three examples demonstrate how to convert a specific use case into a Generalized Use Class. The first two demonstrate how two seemingly different applications may actually have the same video quality needs because they belong to the same Generalized Use Class. The third demonstrates a different resulting Generalized Use Class.

Application 1

Description: Police in---car camera used to read license plates in real time (Figure 7).

The in---car camera must accurately read a license plate of a moving car from a moving vehicle at speeds of up to 70 mph (miles per hour). The application of video in this example involves the police officers in the car, the dispatcher, and command and control. The police car follows the target car, the back of the target car is filmed, and the acquisition of the license plate number is attempted in real time. The alphanumeric character is the target. The motion sources are the target and the camera. The lighting is variable.

Figure 9. Application 1: Generalized Use Class

Application 2

Description: Video shot by a firefighter from a helicopter circling a building fire is used to recognize smoke features from small plumes (i.e., velocity or color) in real time. Command and control personnel on the ground use the video to determine the color and velocity of smoke plumes.

The target is a section of a small plume of smoke, and the motion sources are the target and camera.

The lighting is variable.

Although Applications 1 and 2 demonstrate two different uses of video, the use cases belong to the same Generalized Use Class (Figure 9).

Application 3

Description: Stored surveillance footage is used as courtroom evidence to verify the number of people in a store aisle during an incident.

The end use of the store surveillance is to verify the number of people in the scene. The users are forensic analysts. The sequence of events deals with the people that walk through a store during a period of two minutes. The target is the number of people. The motion sources are the people at walking speed. The lighting is indoor lighting and it is classified as bright.

Figure 10. Application 3: Generalized Use Class

The use case in Application 3 belongs to a different Generalized Use Class, as shown in Figure 10.

IV. Core Video System Components This section provides a summary of the fundamental components of a video system. These components, as shown in Figure 11, provide the foundation for any basic video system. Each component affects video quality (i.e., each component has the potential to alter the video in a way that impacts its ability to be used). The user can use the Generalized Use Class that he/she created based on the selections from Table I to identify performance specifications for the video system components in future guidance material. This section provides information that will help the reader develop qualitative requirements for system components. Future guidance material will provide detailed specifications and additional considerations.

The following sections provide a general description of the components and convey how each can affect video quality.

Figure 11. Core Video System Components

A. Lens Configuration Description

The optical component of a camera system is a lens or series of lenses used to create an image on some sort of media, such as photographic film or electronic means. A lens can be a simple convex surface or composed of a number of optical elements in order to correct the many optical aberrations that arise. A lens may be permanently fixed to a camera or may be interchangeable with lenses of different focal lengths, apertures, and other properties.

Attributes that might affect video quality

Lens Aberration – Lenses do not form perfect images; there is always some degree of distortion or aberration introduced by the lens that causes the image to be an imperfect replica of the object.

Field of View* – The extent of the observable world that is seen at any given moment through the lens.

Focal Length* – Determines the field of view and the apparent size of the objects relative to the image size.

Aperture* – Refers to a lens opening to reduce or increase the amount of light that reaches the image capture surface. The aperture controls the brightness of the image and the fastest shutter speed usable.

Depth of Field* – The range of distances that appear acceptably sharp in the image.

B. Image Capture Image capture is the process of recording data, such as an image or video sequence.

Description

The image capture process consists of converting the information (i.e., light) from a real scene into a stream of information that is suitable for the remaining links via a photographic or electronic medium.

In the case of modern video, the process is modified slightly. The camera is in front of a scene and it has optics (usually just a lens, but it could be a night---vision system). The lens presents focused light to the internal workings of the camera—a projection of the information from the scene. The camera converts the projected information into a stream of electronic data that can support subsequent processing, storage, and viewing. For digital images, the capture process converts light into a digital form via a sensor and digitization.

Attributes that might affect video quality

Resolution at which the camera captures information

Frame rate at which the camera captures information

Fidelity of the colors used

Dynamic range of the recording medium

Number of bits per pixel (digital cameras)

Noise (analog cameras)

Infrared capability of image capture system

C. Processing Description

Processing refers to any enhancement, restoration, or other operation that is performed on a video signal. This could also refer to any processing that occurs automatically as part of a system; for example, ∗ Term is defined in Appendix A: Glossary the processing performed inside a digital camera to convert an image into an image file format. The three main file formats for digital photographs are RAW, TIFF, and JPEG.

Attributes that might affect video quality

Compression* – Also referred to as coding, compression involves electronically processing a digital video picture so that it uses less storage and allows more video to be sent through a transmission channel. Most methods for compression result in a loss of fidelity that is not recoverable. Compression can be used to reduce the amount of bandwidth needed to transmit a video. A user must use a decoder to view a file that has been compressed (or encoded) or else the video cannot be viewed. There are open---source video encoder/decoders that exist on the market; however, there are many proprietary systems that require their own specific decoder.

Digitization – Converting an analog video source to a digital format.

Enhancement for Analysis – Many methods are available to increase clarity for certain parts of the video. Examples are frame averaging, edge enhancement, and color balancing.

Delay – Video images can be delayed, which can result in incomplete or inaccurate real---time decision---making.

D. Transport Transport refers to the effects of moving or copying from one location to another.

Description

Transport and network are terms that go hand---in---hand, depending on the Information Technology Engineer’s preference. This document will refer to this concept as transport. The transport can be wired (including fiber optics), wireless, or any combination of these. The distance of the transport can range from a few feet within a building to the other side of the world. The transport has unpredictable effects on the transmission of the electrical signal between two or more electronic devices.

Attributes that might affect video quality

Available Bandwidth – How much data a network is able to carry affects the speed and size of the video signal that is able to reach the destination.

Network Sharing – Other users on the network may reduce the available bandwidth.

Loss of Data (digital)* – When digital information is transmitted, it is broken into short blocks of data called “packets.” Packets are sent separately and then reassembled on the receiving end of the system. For many reasons, some packets are lost in transmission, causing a loss of some pieces of the video picture.

Loss of Data (analog)* – Noise can interfere with an analog signal (e.g., "snow" on analog TV), permanently obscuring portions of the video field.

Delay – Video images can be delayed, which can result in incomplete or inaccurate real---time decision---making.

E. Storage Description

Video can be used for real---time (e.g., monitoring or tactical) applications or stored for future analysis.

Improperly stored video may be unusable due to loss or degradation of data; for example, improper storage of video would be a critical issue in evidentiary and forensic video applications. Video must be stored simultaneously at a high bit rate and low bit rate to prevent irretrievable data loss. For example, some systems may provide a low bit rate stream for wireless monitoring while simultaneously storing a higher bit rate version locally.

In order to decrease the bit rate, storage is also often preceded by some form of processing. File format can be altered to fit different media, such as coding the video in MPEG 2 for storage on a DVD and playback with a DVD player. A series of alterations or physical custody changes made to a video file is called the “storage chain.” The storage chain should be monitored and documented very closely since almost every change in file format results in a loss of data.

Attributes that might affect video quality:

Physical degradation of storage media over time (e.g., tapes stretching, breaking, or being exposed to magnetic fields)

Physical custody of the media

F. Display Description

To present a true quality picture of video footage captured, the emergency response community depends on a good quality image display unit to aid in accurately communicating information to the end users.

Public safety agencies are increasingly using display functions—EMS for medical diagnosis, fire for research and training, and law enforcement for evidence in the courtroom. Selecting the proper display for the end user’s specific video applications can be as important to achieving the user’s goal for the video footage as selecting the appropriate camera equipment.

Attributes that might affect video quality

"Trueness" of the colors displayed

Aspect ratio used

V. Qualitative Guidance This section provides qualitative guidance for each aspect of the reader’s use class that was determined in Section 3. This section makes general recommendations about how use characteristics and scene content affect component features. It educates readers on the issues involved, tradeoffs, and units of measure associated with the components. This section’s organization mirrors the Questionnaire in Section 3.

A. Usage Time Frame Usage time frame answers the question: “In what time frame will the video be used?” (i.e., will the video be used in real---time or will it be recorded?).

Capturing real---time video

Real---time surveillance requires the integration of cameras and high---quality displays. In defining their needs, users should consider whether cameras will be used for recording a video stream, even if the primary usage is in real---time (i.e., the recording could have forensic or analytic uses down the line).

High---resolution cameras require greater storage because they record video streams and capture minute levels of detail.

The cameras, lenses, or mounts used will depend on whether the user plans to conduct analytics during real---time use of the video system (e.g., using computer processing to aid in the detection of objects, behaviors, or motion). For example, will the video system be used for detection and alert purposes if people approach or loiter near critical infrastructure? Users must take into account factors such as how far away the infrastructure is, how large it is, and whether it is well lit.

Storing video

Videos can be stored for possible future analysis. Proper storage is critical in evidentiary and forensic video applications. For example, if video is stored only at a very low bit rate to save space on the storage media, then data will be irretrievably lost. Those accustomed to storing word---processing documents or digital photos might be shocked at how rapidly video files consume storage space, which can range from single video cassette recorders or a single hard drive to systems with multiple cassette recorders or huge arrays. Depending on the frame rate, resolution, and compression of the video, a user may need to store terabytes of data. One clear implication of this: both the rules for retention within an organization and the records---keeping laws in a particular jurisdiction will affect the amount of storage needed.

Storage is often proceeded by some form of processing, altering the file format to fit differing media, such as coding the video in MPEG 2 so that it can be stored on a DVD and played back with a DVD player.

A series of alterations or physical custody changes made to a video file is called the “storage chain.”

Almost every change in file format results in a loss of data, so the storage chain should be monitored and documented very closely.

B. Discrimination Level The discrimination level answers the following question: to what level of discrimination does the user need to recognize a target of interest?

High discrimination of detail for recognition/identification

If the video will be used for activities such as positively identifying individuals, reading license plates, or detecting small objects, and not for simple motion detection (e.g., did a person cross a boundary?), look for the following features:

High---Resolution Camera – Resolution is measured in megapixels or lines. More megapixels or lines offer a greater ability to capture detail. The tradeoff is that as cameras produce more data, more bandwidth and storage are needed.

Large---Imaging Sensor – Sensor (or “imager”) size is measured in fractions of inches. Large sensors, such as a 1/2---inch sensor, can capture more light than a 1/3---inch or 1/4---inch sensor.

If fidelity is especially important, users should pay attention to other aspects of the camera, such as the manner in which the camera is connected to viewing or storing systems, and the compression ratio of the video signal.

The lens* should offer low distortion and high resolution.

The communications network* needs to accurately transmit all signals. If digital, a high---bandwidth, reliable communications network may be needed. If analog, the communications network should have a good ratio of signal to noise.

Compression* should be kept to a minimum, since a low compression level will result in lower image quality. The amount of storage available and the rules governing retention of the video can determine the amount of compression required.

C. Target Size Target size refers to the size of the object of interest with respect to the field of view.

Capture small or distant objects

Cameras used to record small or distant objects should have:

Telephoto, Zoom*, or Varifocal lens – The degree to which lenses bring distant objects closer or make objects appear farther away (but capture wide fields of view) is called the “focal length.”* Focal length is measured in millimeters. Lenses with longer focal lengths (called “telephoto” or “long lenses”) are more telescopic and can make distant objects appear larger to the camera’s sensor, so that images will be sharper. The drawback is that the longer the lens is or the more it is zoomed in, the less field of view* the camera will capture. The advantage of a varifocal or zoom lens is that it can act as a wide---angle lens to capture a wide scene and zoom in on small or distant objects. One tradeoff of long lenses is their susceptibility to capture less light, resulting in less functionality in dark environments. Furthermore, telephoto lenses that admit a large amount of light are usually large and expensive.

High---Resolution Camera – Resolution is measured in megapixels or lines. More megapixels or lines offer a greater ability to capture detail. The tradeoff for this ability is that as cameras produce more data, more bandwidth and storage are needed.

Large---Imaging Sensor – Sensor (or “imager”) size is measured in fractions of inches. Large sensors, such as a 1/2---inch sensor, can capture more light than a 1/3---inch or 1/4---inch sensor.

If faithful capture or recognition of these small objects is especially important, users should pay attention to other aspects of the camera, such as the way the camera is connected to viewing or storing systems and the amount that the video signal is compressed:

The lens* should offer low distortion and high resolution.

The communications network* needs to accurately transmit all signals. If digital, a high---bandwidth, reliable communications network may be needed. If analog, the communications network should have a good ratio of signal to noise.

Compression* should be kept to a minimum, since a low compression level will result in lower image quality. The amount of storage available and the rules governing retention of the video can determine the amount of compression required.

D. Motion Motion refers to the motion in the scene of interest (e.g., background, target, camera).

Camera or objects in motion

The ability for a video system to clearly capture moving objects or to capture scenes if the camera is in motion is determined by several factors. Systems are better suited for this situation if they have:

Fast Shutter Speed – Shutter speed is the length of time that the camera sensor is exposed to the image it is capturing for each of the still photos that comprise the video. If there’s enough light to do so, it’s better to have fast (or “short” or “high”) shutter speeds (measured in fractions of seconds) so that while each image is being recorded, the scene being recorded on the camera has less of a chance to move and thereby blur the image. One way to decrease the need for the environment to be bright is to use a wide---aperture lens.

Wide---Aperture Lens – The ability for lenses to admit more light is measured in f---stops; the smaller a lens’s f---stop number, the more light is allowed to pass through the lens. The advantage of a smaller f---stop number is that when more light hits the sensor, the camera uses shorter shutter speeds and takes more pictures. However, this come at a price because lenses with small f---stop numbers require wide---aperture lenses, and those lenses have difficulty keeping both near and distant parts of the scene in focus at the same time (a quality known as depth---of---field*).

Wide---Angle Lens – If the camera or object is moving, it’s easier to keep the camera on the object if the lens is capturing a wider scene. This ability of a lens to capture a wide scene is a function of its “focal length,” measured in millimeters, with smaller focal length lenses capturing wider scenes, but making them appear smaller or farther away.

Large Imaging Sensor – Sensor size is measured in fractions of inches. The sensor is to digital cameras what film is to conventional cameras. Large sensors, such as a 1/2---inch sensor, can capture more light than a 1/3---inch or 1/4---inch sensor.

E. Lighting Level Lighting level is an obvious aspect that impacts the selection of video equipment and refers to light available to illuminate the scene of interest.

Low---Light Situations

The ability of video systems to work in low light is measured in their “lux” rating. Lux is a measure of the amount of light present; cameras that have a lower lux rating can work in darker environments.

As with high---motion applications (described in Section V. Qualitative Guidance, D. Motion), when selecting lenses for low---light applications, try to select those with larger apertures (lower f---stop numbers). The larger opening will transmit more light to the camera’s sensor. This will give the added benefit of reducing motion---blur, since each image will be captured more quickly. In other words, users will be able to use faster shutter speeds.

Camera considerations for very little or no light:

Black---and---White Camera: Most black---and---white cameras are better suited for low---light environments than color cameras.

Day/Night Camera: This tradeoff may not be easy, but the increasing availability of “day/night” cameras designed for color during the day and black and white at night increases the number of options for wide---ranging lighting situations.

Infrared Camera: There are also cameras available that work in zero lux, or with no visible light. This is possible because these systems are sensitive to infrared light, and they use a form of infrared lighting that is not visible to the naked eye.

Changing---Light Situations

If the video system will be used in environments where the light can fluctuate from bright to dim, be sure to consider “auto---iris” lenses. These lenses automatically adjust their sensitivity in changing light situations so that the lighting level of the video appears as even and constant as possible.

VI. Ongoing VQiPS Efforts OIC, the VQiPS WG, and PSCR are continuing their collaborative and complementary efforts to identify standards development guidance, video quality performance qualifications, and other information that is important to those procuring video systems. OIC will continue to publicize the benefits of using this guidance document with local, state, and Federal emergency responders to build ongoing support throughout the emergency response community. Future rewards of this continued outreach will be new stakeholder champions within the VQiPS community and adoption of standards by industry and standards development organizations.

OIC continues to participate in and provide support to the VQiPS WG to deliver future products, including compiling an inventory of existing standards and specifications that address various components of the video system core components for specific usage scenarios (or Generalized Use Classes). With this information, the emergency response community will be able to match particular needs to technical performance specifications and standards.

Funded by OIC, PSCR is working with industry and the emergency response community to develop a set of application---independent usage scenarios and conduct lab research to develop video quality performance qualifications. PSCR is developing performance specifications that will be available to the emergency response community upon completion.