Big City PSIM Claims Reviewed

By Carlton Purvis, Published Nov 21, 2013, 12:00am EST (Info+)

The U.S. has one of the highest crime rates in the world. Within the U.S., Baltimore is one of the worst and its need for security and surveillance is particularly acute. IPVM recently participated in a tour of the city's command center during Secured Cities. In this note, we examine the system and some of the more unrealistic claims made.

System Overview

********* ***** *** ****** '********* [**** no ****** *********].' *** ****** ******** 635 *******, ** **** ** ** 2006. ** ********** ******* **** ******** *********** of **************, ********* ********** *** ******** Security ***** ***** ********. 

********* ********* ***** **** ** **** Citiwatch:

  • ************ ** *,*** *******
  • ********* *,*** ***** ******** **** ********** criminal ********
  • ********* *** ******* *** ******* ****** (including *********, ***** *** ******* - though *** ****** ****)
  • ********* ** ******* ******* ******

*** **** ** *********'* ************ ********* can **** ** **** *****, ****** media ******* *** ************ ****** ************** using * ****** ****. **** ********** with ****** ** *** ****** *** other ******** *** ***** *** ******* an ***** *** ** **** ****** arrests *** **************. *** ****** ******* **** easily ************, *** ***** *** * number ** **** ** **** ** video. 

***** *** **** *** *** ***** photographs, *** ******* ****** *** ***** the **** ** *** ********** *****, with *** **** ********** * ***** screen ******* *** * ********** ***** and *** ***** **** *** ******** and ***** *** ** ** ** operators. ** **** **** ** *** tour, ***** **** ** ********* (******* police ********) *******. ****** **** *** eight *****.

**** ** *** ******** ******** ******* **** ** *** ** ***** on *** *** ******:

*** *** **** ** *** ****** is * ****** *** ********* *** click ** *** ******* ** * certain ********. *** ****** ****** ** a ********** ** ************** ******. *** bottom **** ** * ******* ****** camera. *** ** *** ***** ****** the ******** ** * ******* ******** map ***** ******** **** * *** browser. *** *** **** ** *** screen ** * **** ** ******* and *********. 

Questionable ******

***** *** * *** ****** **** by ********* ******** ****** ********** ****** Hood ****** *** **** **** ******* some ********:

Instant ****** ** *******

** * ************ ** ******* ****** the *** ***** ** *** ****, he **** ***** ******'* ******* ***** be ****** ** ***************, *** ****** the ****, ************* ***** ***** ****** a ****** *******, *** **** ** took *** ****** ****** *** ****** with * ******* **** / ****** displayed *** * ****** ** ****. "Instant ******" ** ********* **** "****** in *** *******," ********** ** ** emergency ** ********** ** * **** security ********.

Capturing ******* ** ****** ***** ********* ** *****

*** ******** ****** ** ***** **** a ******** ******** ** ***** ** night. * ****** ** *** ***** highlighted **** ** **** *** * bullet ****** ** **** ** * camera ********* ** *****. **** ** looked **** ** ** *** * cloud ** ********* ******* [**** ** ****** available], ***** ***** ** ******* ****** than * ****** *** ***** ****** long ****** ** ** ******** ** camera. ** ***** ** ** ********* lucky **** ** ****** * ****** with * ****** ********* ** *****, especially *********** ** *********** ** ***** take ** ***** ****** ** **** light ** ******* * ***** **** of * ****** ********. 

No **** *** ******

*** ******** **** **** ********* *** eliminated *** **** *** ****** ********.

"** * ******* **** ****** * corner, ** **** ***** ** ** them ** *** **** ***** [**** the *******]," ** ****. **** ****** good ** ******, *** ****** ** many ******* ******** ***** **** ******** rather **** *********** ****. ****** ****, 635 ******* *** * ********* **** ********* ****** ** ****** such ********** ** ********.

Operators *** ********** ******** ****** ** ********The cameras are monitored around the clock. Hood said operators will routinely alert police to crimes before dispatchers even get a 911 call. Hood did not provide any hard numbers on how many times police were dispatched based on something an operator observed, but said it happened "all the time."

** ****** **** ************ ***** ******* of ******* ****** **** ******* ** riots, *** **** **** ** *** videos, ***** ** ** *** ** tell **** ********* *** ******** ** key ** ** * ******* - a **** ** ***********. 

**** *** ****** *** **** ******* is **** *** ********* ** **** some ***** ** ********* **** * person ** ******** * ********* ****** and *********** ********** ********. ** *** *****, the ******** ******* * *** ****** ************ near * *** ** ****** ****. The ******** **** *** ****** ** follow *** *** ***** ** ***** a *** ******, **** *** ****** dispatched ** *** ********. ** ******* video ** ******** ******* **** ***** like * ********* ****** ***** * teen's ******. *** ******** *********** * stop **** ****** *** ******** **** a *** ** *** ****. 

Caught ** ** *** ****

* ***** **** **** ** **** that **** *** ***** *** ********* get ****** ** ** *** **** when **** **** ** ********. ****** the ****** ** ****** *** **********, but ****** ** **** ** *** to ****** *** ******** ******* ** camera ** * *******. *** ********* slogan ** "*******," *** ********* ****** are **** ** **** ***** ****** with * ***** ** ****. 

No *********** ** ************

*** **** ******* ** *** ************* of * **** **** ** *** already ** ***** *** ****** ***** what ** **** ** *** ** up. *** ******* **** ************ **** a ****** ** ******** ********* *** Maryland *********** ** **************, ********* ********** and ******** ********, *** *** *** using ********* *******, *** ********** ******* *********. A **** ******-** ** *** ************ would ** ** *********** ******* ** facilitating ************* ******* *** ********, ***** probably *** ****** ********** ** ****.

Comments (13)

This is a good article. It shows that there is still a large gap between TV/Filme and the reality of real surveillance systems. Its very hard to get an audience with the implementers on how they went about it. They tend to hide behind the security screen "I cant tell you that otherwise we would be exposing our weaknesses"

Very interesting article. I wonder what the response time is from when a surviellance operator confirms a suspisious suspect to a police response and possible arrest. Also, the section pertaining to a camera capturing a moving bullet at 30FPS had me laughing. The Citiwatch slogan "Beleive" sure is befitting.

The response time is pretty fast. Operators have direct radio contact with police.

People rarely want to admit mistakes in some things like purchasing decision and will try to put on a best face, especially when the public is there in your face judging value.

But overall I don't think it's a bad system. It certainly didn't need the embellishments.

There are only three statistics I care about, and I only see one of them: participation in arrest (which we see), participation in prosecution, and cost per succesful prosecution.

Arrest is nothing without prosecution, or at least an attempt at prosecution. And the NYPD's stop and frisk policies show that you don't need to have a lot of fancy, expensive technology to arrest everyone who looks suspicious or walks funny, just a lot of cops.

How can we evaluate the effectiveness of this system without these statistics?

A comprehensive study of the Baltimore PD was just released last week. At a cost of $285,000, the report discusses certain aspects of CitiWatch. As one of the most violent cities in America, I think one has to ask "What would the city look like without CitiWatch and how can CitiWatch become even more effective?". I participated in this tour and definitely learned a few things to apply in my agency.

Mark, thanks for sharing that. What cost $285,000? The report or the system? If it's the report, that sounds like a lot. If it's the system, that sounds way too low :)

John

The consultant's report cost $285,000. Not sure about the entire CitWatch system.

"In a presentation at Secured Cities the day prior to the tour, he said other agency's cameras could be called up instantaneously, but during the tour, instantaneous often meant within a couple minutes, the time it took the system access the camera with a loading icon / symbol displayed for a period of time. "Instant access" is different than "access in two minutes," especially in an emergency or responding to a live security incident."

I find this problem surprising... we have delivered many interagency projects (e.g. safe city, defense) and we have yet to encounter this kind of issue. Did they provide any explanation on why this delay occured? Maybe there a "human in the loop", e.g. another agency's operator must first manually authorize the transmission of video to Citywatch?

Christian Laforte
CEO, Fortem

This is in the basement of the Baltimore command center:

It's a good question. Carlton and I talked about this before publishing. The cause was not revealed. Carlton can try checking, but Vidsys is long on marketing and short on real explanations.

Thinking about it, there could be plenty of valid reasons to use a man-in-the-loop approach to authorize video transmission from one agency to another. Legal restrictions, minimizing upload bandwidth, preventing unauthorized information leak, etc. My bet would be on security personnel that routinely watch the public, but are afraid of being watched by other departments and being held responsible if something goes wrong. That kind of double-standard is almost the norm, and it's holding back a lot of potential cost savings and improvements in operational efficiency.

We are just finishing one interesting safe-city project that isn't held back by this kind of organizational fear, involving 6 agencies that want to share cameras and analytics. We had to extend a lot of our permission features, e.g. to support complex new rules to automatically grant temporary permissions to some cameras based on opened alarm type, escalation, location, etc. Some agencies have policies that restrict archive vs live feed sharing. Fun stuff.

Yep, I'll see if I can find out.

Baltimore in the past said the operating cost for this system was $1.4 million annually and is now requesting an additional $700,000. The city also increased it maintenance contract to $8.2 million to account for "platform integration and a limited expansion of the system." All this after a review found the system had severe deficiencies.

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts reporting, tutorials and software funded by subscriber's payments enabling us to offer the most independent, accurate and in-depth information.
Loading Related Reports