From his office in the Chicago (Illinois) Police Department's 16th District Tactical Unit, Lt. Joe Schober reviews the usual array of case and arrest reports collected over the past 24-hour period. But in recent years, a few new items have been added to the information he must review. This new information comes in the form of video sources from retailers and gas stations, and these videos contain footage of robberies or thefts.
Two years ago, it was considered unusual if a district or area received more than one or two videotapes or disks a week. Today officials might receive two or more videos per day. The influx of video recordings requiring viewing and processing created a variety of problems. All video sources coming into police custody must be inventoried and subsequently viewed by a special sections unit. The equipment the section traditionally used was quite sophisticated but not intended to deal with a variety of video sources. Additionally, the manpower allocated for video workups remained the same while the number of incoming videos increased. Determining which videos to review first also became a concern. Videos containing information on violent crimes receive top billing. This translated into increasing delays, and as such delays multiplied, the opportunity to successfully pursue criminals recorded on video dwindled.
But the utility of video cannot be underestimated and thus Chicago had to find a means to address these problems.
"There's a big difference between the offender descriptions provided in the identification boxes of a case report and the description that can be seen when reviewing video of an actual crime," explains Det. Jim Adams of Chicago PD's Area Five Detective Division.
A major benefit of having good video is that it can be repeatedly reviewed, diminishing the chance for confusion. For instance, victims are often so traumatized during the commission of a crime that their acuity, even within minutes of the event, can be drastically and detrimentally affected. If the suspect brandishes a gun or a knife, victims often focus on the weapon rather than the offender. Descriptions of height, weight and age are frequently so subjective that two people witnessing the same incident may give police entirely different accounts.
Furthermore, video helps authorities address deliberate attempts by criminals to confound identification, such as ditching articles of clothing immediately after a crime. Because most people are aware that gangs often display or wear colors associated with their affiliation, more generic clothing has become the fashion-of-choice among gang members. Consequently, offender clothing descriptions often include a suspect wearing, for example, a dark hooded sweatshirt, or more commonly, a white T-shirt and jeans. It is not unusual for a group of offenders, intent on committing street robberies or muggings, to flood an area with people in similar dress in order to confuse victims, witnesses and the police after a crime has been committed.
During the course of an investigation, Adams contacted Procom, a local business in the 16th District that specializes in surveillance equipment, and met with Paul Pustelnik, one of Procom's senior members. After discussing the problems encountered in the timely processing of video recordings, Pustelnik suggested Adams use some of Procom's equipment to see if any of these problems could be resolved. With this equipment, Adams tested sample videos that the forensic unit had experienced trouble opening or from which no useful footage could be obtained. Within minutes, Adams not only opened the problem recordings but produced reasonably good quality stills.
Procom provided the district with equipment and software for a trial period to allow officials to assess its usefulness as an investigative tool. After using the equipment for just over a month, more than a dozen arrests and more than a score of clear-ups were directly related to the use of this video enhancement equipment. In one instance, three arrests were made within hours from the original video of a pawn shop heist. While that video was being processed, district tactical officers present at the scene identified the subjects in the video as known gang members. Officials presented a mugshot photo array to the victim, who positively identified the offenders, and police later arrested them.
Procom video enhancement products enabled the district to easily develop other applications that allowed still images, captured from video, to be electronically displayed during roll calls, sent to headquarters for crime watch alerts or bulletins, or reproduced for posting on unit bulletin boards.
Word of these successes informally spread through the department, and Schober soon began receiving recordings from other department groups. This indicated to him that what the 16th District was accomplishing might be of value to others.
In June 2006, Police Superintendent Phil Cline visited the 16th District to personally witness the results obtained through Procom's approach to video enhancement. After further demonstration at police headquarters, he directed steps be taken to implement the system in all five detective areas.
Procom and codecs
To a large extent, the success Pustelnik's equipment demonstrated came from Procom's knowledge of codecs.
Codec is the truncated version of two terms: code-decode (sometimes referred to as compression-decompression). Codecs provide the encryption for viewers that allow viewing of a specific video format. Often codecs are built into existing system platforms. Microsoft's Multi Media Player (MMP), for example, contains a number of codecs that allow video viewing.
Codecs are used for video and audio files, and can be identified by their extensions, which appear after the file name. Sometimes a video file codec may have to be converted in order to view results on a viewer. Audiophiles know there are times when a codec for an audio source may have to be converted to play on an MP3 device, such as an iPOD, rather than on a CD.
At the current stage of audio and video development, there is no "standard" for all files, which means there are several hundred codecs to choose from in order to obtain desired results. Fortunately, most files (probably 80 percent or more) may be opened with a few strong codecs that are generally available for download on the Internet.
What's inside the "box"
The equipment that will be described shortly didn't just come out of a box; the Chicago PD started with equipment deemed best for the task at hand and added or subtracted from there.
The 16th District's basic system begins with a dual core processor and RAID architecture. Video files can be large and take up a lot of space on hard drives so two 300-GB drives were employed. RAID 0, a mirroring technology, allows the drives to work in tandem to create the equivalent of a 600-GB hard drive and builds in a security measure to save data in a single drive failure. The memory was 2-GB SDRAM, although more is better. This system includes a DVD player, DVD-Writer, and seven reader ports to accept standard USB 2.0 flash drives and a variety of camera video memory cards, which allow for data input and recording.
A 128-MB video card was installed to handle the high rate of information processed from video sources. The Chicago PD had used an ATI card, a good product, but one that conflicted with some of the software being used. Officials found a NVIDIA NX6200 card used with a Pinnacle capture card worked best for the department's purposes. Additionally, officers used an IEEE 1396 (firewire) card to transfer video directly from video cameras.
A time lapse video recorder for VHS (analog) tapes also was used. This device doubles as a VHS player and a "time base converter," an option that can be expensive. Time base conversion stabilizes and adjusts erratic video signals. The recorder is currently useful but may be considered optional as video sources trend toward digital.
The most obvious feature in the design of this system -- the one that gets the most immediate attention -- is the two 19-inch flat-screen monitors on the desk. Dual screens allow the user to view multiple applications without cluttering a single screen. For example, the user might view one screen, which contains the original image, while working on an image enhancement with different software on the other screen.
At this point purists might ask, "Where's the SDI (analog to digital) converter?" or "How are you going to handle multiplexing?" and "If you're going to take a nonlinear approach to forensic analysis, you're doomed!" The most honest and straightforward response to these statements is that necessity is the mother of invention. While sophisticated equipment intended to resolve some of these issues can be purchased, this equipment is expensive. Chicago's solution involved an ad hoc approach to needs, using equipment and capabilities it already had.
Pinnacles' Studio 10 and Adobe PhotoShop SC are the principle software used in the 16th District design. Studio 10 is used exclusively for VHS tapes to capture sections of video related to an incident. Once captured, the event may be saved as an avi file that can be reviewed on any number of viewers including MMP. Additionally, Studio 10 allows the user to capture still images from the video by using tools provided from the program's pull-down menu. The stills are saved as JPEG files, which are suitable for PhotoShop enhancement.
Chicago's enhancement process
Acquiring the recorded event from a crime scene is often accomplished with the help of personnel trained in downloading video files from the victim's system. It is a good practice to witness and note the steps used in this procedure as many source machines employ variations of the same process using different steps and passwords. It should be noted that digital sources copied to disk are not the original recording which, in fact, stays with the machine. Analog or VHS tapes are the original source and, when possible, should be duplicated for use in video enhancement. Using secondary sources, or copies, of the original source has important implications for the evidentiary chain.
Captured stills enhancement from DVR or CCTV systems is slightly more involved. The instructions provided here apply to systems using Microsoft Windows XP. The process initially involves inserting the video source disk into the machine's DVD player and waiting a moment. An information screen should appear offering users a variety of options related to the inserted disk. It is important that the disk is recognized here, which is a positive first step. Users then close the information screen and right click on the start menu. This prompts a display to the left, which shows the computer's contents, including what is in its DVD drive. A double click on the DVD icon, and the disk's contents display to the right of the screen. The contents provide information related to the types of files on the DVD. Most of the files will be named or numbered and include extensions that identify file type. The disk also may display the viewer used to open the files. A viewer will appear as an icon, which is distinctly different from the other files. If this is the case, users simply double click the icon and the viewer codec loads to the system. Once installed, users open the viewer and select the browser option to locate the DVD file for review.
In some instances, although the files are evident, there is no viewer icon. Here users must open Windows MMP and attempt to play one of the files from the disk. If the file won't open, the user will be prompted that help is available via the Internet. Because the file extension type is known, users can seek assistance according to the extension or download other viewers, such as Win Amp, that have strong codecs capable of opening several popular files. In most instances, downloads are free. With a little practice, this exercise will prove invaluable to resolving codec issues. And once four or five codec viewers have been installed, they typically can handle the majority of video sources encountered by police.
Like Studio 10, several viewers have capture options that allow users to take still shots from the video source, though this is not always the case. Adobe's PhotoShop, however, offers a viable solution that can be accomplished in two sets of steps. First users must locate the print screen button on the keyboard display. During the play of an open file on any viewer they may press this button and hold the screen image into memory. Now they open PhotoShop and select "New" from the File pull-down menu. A screen will appear and users simply press "OK." The next screen that appears will be blank. Users then go to the Edit pull-down menu and click on "Paste." This places the captured screenshot in the blank space but users are only concerned with the photo image display on that screen. Users then choose the dotted square icon from the PhotoShop toolbar to the left and drag the square from corner to corner over the section to capture. Users then return to the Edit pull down and select "Copy" and repeat the steps in italics above. This image should be saved in a folder related to the event as a TIFF or JPEG file. The file naming conventions selected are ultimately up to the department, but short names, numbered in sequence, are the easiest to remember.
Another available option involves the use of Microsoft Film Maker (SP2), which is a robust program that can provide both video and still images from CD or DVD sources. After opening the program, users select the "Import" option under the File pull-down menu. If the file can be opened by this program, Film Maker breaks the film file into smaller sections that display on the program's viewer. The program slider allows users to move freely back and forth along the frame's timeline. The Service Pack 2 (SP2) version also has a small camera icon beneath the viewer that allows users to take still shots (JPEGs) from the video source. Again, users must follow a regular naming convention when saving the captured stills.
Now users have a set of captured images that will probably need to be enhanced. Enhancement deals with rendering a captured image to a form that is clearer or more recognizable without changing the image itself. The task is to make the image bigger, better and brighter. To do so, users open the captured image in PhotoShop. Images may be in color or black and white. At times it may be best to eliminate noise from the image by eliminating color. This is done through use of a desaturation option under Image Adjustments. Another reason for making this change is the printer. If users are working with a black-and-white laser printer, the results of the enhancement will appear clearer if the image is already in grayscale format.
Next users go to the Image pull down again and select "Adjustments." Although there are a variety of options, users are principally concerned with two: Brightness/Contrast and Curves. Curves presents a graphical representation of the image with an extended line. Click on the middle of the line and pull the line to the left or right -- parts of the image will darken or lighten. Clicking on the line above or below the center affects other sections of the image in the same manner. Anytime the results of the last instruction are unsatisfying, they can be undone in the Edit menu. If this has helped, users then close the graph and open the Brightness/Contrast option. Brightness/Contrast allows users to play with the lighting and shadows within the image. The results of all changes can be seen immediately within the image. The "sliders" in this option allow users to return the image to its original state by setting the point indicators to zero. Once finished, users close out of the option and save the image.
PhotoShop allows users to save images in a variety of formats. Because image adjustments have been made, it is advisable to save it as a JPEG for a couple of reasons. First, the file takes less space on a hard drive than a number of other formats. Second, JPEGs are easy to import into other applications or e-mails.
The saved JPEGs are used for a variety of purposes and can be arranged in a number of documents. For several reasons, the Chicago PD finds that placing images on PowerPoint slides enhances the variety of uses for captured stills. In PowerPoint use the Insert pull down to retrieve a picture from a file and place or resize the image, or combination of images, anywhere on the slide. PowerPoint also allows users to easily add and move text to the slide so the Chicago PD places case numbers, addresses, type of crime and notes related to the stills on the slide(s). The finished presentation can be e-mailed to other units who now have the ability to lift images from the slide and use it for alerts or in-house flyers. Most Chicago districts have large monitor displays in the roll-call rooms and the presentation can be displayed for officers before their tour. The main purpose of Chicago's Video Enhancement Unit is to provide the best rendition from a video image source that recorded criminal activity and provide that rendition to officers and investigators in a timely fashion.
A complete success
While the most recent version of the process described in this article has been in place for less than a year, it has contributed to the arrest of 63 offenders and to clearing 131 cases related to subjects involved in armed robberies, burglaries, thefts and damage to property. The majority of these arrests involved felonies which transpired in front of surveillance cameras at gas stations, convenience food marts, banks, retail stores, drug stores and fast food restaurants. Many of the captured stills assisted other forensic groups who were able to review where an offender had been and what was touched prior to processing a crime scene.
Rendered images of video sources has proven an effective tool for law enforcement in Chicago. Certainly there are more elaborate, and expensive, video enhancement systems available for more sophisticated needs. But for a majority of departments looking for a video enhancement solution to address the growing boom in surveillance sources that come into police custody, this is a great starter kit.
Tobin Hensgen is a 20-year veteran of the Chicago (Illinois) Police Department where he has served in a number of capacities. He holds an MBA in information technology, a master's a degree in public administration, and is currently completing a Ph.D. at Loyola University. Hensgen has contributed material to professional journals and magazines, and his books on managing information are available in more than 200 universities worldwide.