Advice for making your own web cameras page


Overview

How the "Hoodoo Butte pictures" page is created is described in a separate "about this page" document. This page is about the design and installation of the hardware that is needed to make it happen.

You never know when someone may want to access your page, and you may be surprised at the variety of their interests. A web page with stale information or dead links is at best an annoyance, and may mislead an unwary visitor. A reliable web page begins with a reliable hardware installation. It is not easy, but it is do-able.

Reliability is a result of good design and installation practices. It starts with getting something to work under ideal conditions. It should continue by trying to anticipate and defend against the effects of all non-ideal conditions. "All" covers a lot of possibilities, so expect a design to have weaknesses. After you have done your best to anticipate the foreseeable problems, nature and time have a way of revealing design flaws. To find them, all you have to do is watch the system run.

Problems with a design after installation should be viewed as the result of inadequate design. Typically due to inexperience, you may have failed to anticipate a situation, chosen to ignore it, or implemented an inadequate defense. Emergent problems are to be expected, and so should be the effort to fix them. This doesn't have to be a full time effort, and a system becomes more reliable as the frequent problems are eliminated.

In defending against non-ideal situations it is helpful to first of all know what they are. For each situation you should have an idea how often it happens, and what affect it may on the installation. The effects may range from nothing or mild degradation to temporary inoperability or permanent damage. Most effects can be reduced or eliminated by a more robust design, usually at a cost of increased design complexity. You may need to accept that for some events the best you can do is simply survive to operate another day.

Realize that a design that's perfect for one party may not be for another. You might not have the same problems, or need the same solutions. Available skills, resources (time, money, people), and technologies tend to determine which options are reasonable for you. What is being offered here are the "Lessons Learned" from this installation.

If we were to build a new site, or make major changes to the current installation, the following topics would be considered.


View Selection - What are the elements of interest?

Before you start buying cameras and lenses, realize that you can't make wise choices until you know what you need. This can be done in an orderly manner, and it starts by thinking about who will be looking at these pictures and what is of interest to them.

Different users may look at the same picture and see different things. A radio dispatcher may concentrate on activity in the foreground. Weather observers may look at the sky, and potential visitors may check out the scenery in the background. These "elements of interest" may be different for each user, yet it's possible for a single view to include them all.

The person or organization sponsoring your efforts should have particular elements of interest in mind to serve their intended audience. Their interests must be satisfied, or at least listed, prioritized, and met as best possible. (They may want more than they can afford.) At the same time, with little or no additional expense, you may find that the interests of other users might be served as well. Try imagining who they may be, and what they would find interesting. You may find a new type of customer in these users.

We can usually build a list of possibilities which easily exceeds our budget. Now it's time to prioritize and separate the "must" from the "want" items. You may want to start simple and expand later, after you have some experience with success. Don't lose the "want" list, there should always be a place for your dreams.

Some views will be selected because you need them. Significant investment may be required in order to get that view. Other views may be "targets of opportunity", where the infrastructure is largely present and it would be easy to add another camera. As you work to meet your "Must" priorities, consider what else may be possible for little additional expense.

Long distance visibility is a frequent casualty of the weather. If possible, include elements in a scene both near and far. In that way, when it gets really foggy, you'll have something to see besides the inside of a cloud. Even a nearby tree can be an assurance that when visibility improves, the view will return.

Respect the privacy of anyone who may be in the field of view for any reason. Exterior views are generally uncontroversial. Cameras inside buildings can be unwelcome, as most people don't like being "watched".


Site Selection - Where to place a camera

Now we are looking for places to mount a camera. It's time to visit the proposed sites and look around. Take a camera, measuring tape, and notebook, so you can record basic information about where things are located and get some dimensions. Look around and try to identify alternative sites, in case your first preference turns out to be impractical. Keep the "must" and "want" elements of interest (developed previously) in mind as you look around.

For each alternative, try getting a picture of the view from that location. This will help in site selection, and possibly serve as the first picture in your prototype web page. Other pictures may help with the design process, by noting where potential useful (or interfering) structures may be located. As you look around, realize that a site must meet a few fundamental criteria:

  1. View
    • Does the view include the "must" elements of interest?
    • Are there other interesting views?
  2. Safety
    • Does it avoid creating additional safety hazards?
    • Staying "out of the way" is a good idea not only for the safety of others, but also for the survivability of your installation. People may blunder into it by mistake, or heavy machinery may trample or bash it.
  3. Agreement
    • Does the owner of the site understand what you are proposing to do, and do they agree to allow you to do it?
    • This may take a few discussions, as you may not appreciate the extent of the final site design until later in the design process.
  4. Electrical power
    • Can you plug-in, or bring it from nearby, or will you need to generate your own power from sun, wind, propane or other energy sources?
  5. Network connection
    • Can you connect to your network using wire or wireless methods?
  6. Room
    • Can you fit all you need into the site?
If these criteria are met most other shortcomings can be addressed in the "Site Design" phase.

Making use of existing structures for support and protection (like sign posts, towers, and buildings) can relieve you of having to build the equivalent yourself. This can save a lot of time, money, and effort, and minimize the visual clutter your efforts will bring to the scene.

When you share a resource like a building, consider your impact on the other users. Antagonizing them may get you invited to leave, or simply treated with contempt. Keeping hardware running often takes good will and cooperation, so try to locate your equipment out of the way and minimize resources (like electrical outlets) that you will be using. This can help protect you from accidental damage, as the other parties are not necessarily concerned with protecting your best interests.

This initial site survey might be all you need to start developing a site design. Dimensions and locations of support posts, towers, windows, wall space, and lengths of cable runs would all be handy. Determining the angular width of the desired field of view may be needed to select a suitable lens. A stick held at arms length can be used to estimate the angle (to be measured & computed later), or draw a map of the site and include sight-lines for the field of view..


Site Design

If you think a camera site will be simple, you probably have overlooked something. As mentioned before, design begins with a basic installation which works under ideal conditions. Design should then continue, attempting to anticipate and eliminate (or mitigate) problems due to all possible non-ideal conditions.

Notice that the problems aren't limited to "forseen" conditions. What you can imagine may be only a small part of what nature eventually throws at it. People, either disinterested or malicious, can add yet more challenges.

"Non-ideal conditions" may affect your installation in one of these ways:

  1. It has no effect
    - This is the objective of good design.
    - Designing a problem "out of existence" may be required to achieve this.

  2. It may degrade operation
    - Still working beats the next alternatives . . .

  3. It prevents operation, but does no damage
    - Surviving to operate another day can be a reasonable strategy.
    - The ability to recover without manual intervention would be good, as needing to re-visit the site can become annoying.

  4. It quits working due to damage
    - The design failed.
    - This can be a slow and expensive way to get an education, and it's bad for reliability and reputation.
Being able to anticipate unusual conditions generally involves challenging every assumption you may make about an operating environment. Operational experience is useful, but that's usually what we don't have at this point in a new project. Here are a few examples of non-ideal conditions that we have seen: Minimizing the impact of events such as these should be a goal of the design effort. Consider whether the system will be able to automatically restore itself to normal operation. Manual intervention may eventually become annoying, or it may be impossible under some circumstances, so the ability to gracefully handle a problem time after time may be worth considering.

Not every problem requires a complex solution. Simple can be good enough, cheap, and reliable, but it may not be the first thing we think of when trying to solve a problem. (NASA invented a nitrogen-pressurized ball point pen for its astronauts to use in space. Cosmonauts used pencils.)

Almost every one of the above examples has happened, and could (but will not) be described in annoying detail. In the case of power line surge protection, we once lost a computer power supply due to someone using a circular saw plugged into the same outlet. Surge protectors need to be located between what is creating the problem and what you are trying to protect. Don't make your protected power outlets accessible for use by other users.

Frequently used solutions for some of the above commonly encountered problems:


Site Installation

Planning and preparation can make it possible to finish an installation reasonably quickly and successfully. It involves more than just hauling stuff out to a site and starting to put it together. The penalty for not doing this phase well increases as the installation site becomes ever more remote.

Planning should start with the making of four checklists:

  1. Tasks - What needs to be done, and in what order?
    - e.g. Install window, build shelf, mount stuff, connect it, turn-on, test operation, ...
  2. Equipment - What are the major items to be installed?
    - e.g. Camera, video encoder, network interfaces
  3. Supplies - What else is needed to make the "equipment" operational?
    - e.g. Tripod head (camera mount), outlet strip, low voltage control & power wiring, black-out curtain, shelf brackets, . . .
  4. Tools - What is needed to put it together and check it out?
    - e.g. Drill, screwdrivers, multimeter, video monitor, laptop computer, scissors, . . .
    - This stuff will not be left at the site after installation is done.
Don't expect to complete each list in one try, this effort will take time and some work to capture all of the details. Try starting with the task list, and consider what needs to be done and in what order. The list of tasks will probably span most phases of the typical project: Now consider, for each task, what needs to be in the other three lists to successfully complete that task. As you think about the entire job you may discover additional tasks and identify their related resources.

A complete equipment list is important because forgetting a major item will mean that some part of the site design isn't going to work.

Supplies are the small stuff like wire, nails, wood, and conduit fittings that help mount things and connect them together. Forgotten supplies may not prevent an installation, but they can slow you down and affect the quality of the workmanship.

Tools, like supplies, can be critical or just nice to have. There is usually a "right" tool for every job, and likely many "wrong" tools that could do in a pinch. Using the wrong tool may adversely affect time, effort, and quality of workmanship. Be especially vigilant for tools which are infrequently used, as they are most easily overlooked.

These checklists are essentially another design effort, so a good "design" needs to anticipate more than just installation under ideal conditions. What additional issues might you encounter and need to handle?

Before taking equipment into the field, consider assembling and testing it in a location conducive to troubleshooting. New equipment may not work, or cooperate with other equipment. The place to find and fix these problems is where debugging is most easily done. Reading user manuals is much more pleasant in the shop, and you may discover issues that are best handled before going into the field.

Getting successful operational and troubleshooting experience is also handy. The experience you gain at the test bench may also help fill-out your tool and supply checklists. Ideally the field installation will involve little more than mounting the hardware, making connections, and verifying that it works. Take the User Manuals, they may be handy in case of difficulties.

The actual installation can be a bit like an expedition. Use your checklists, but stay flexible. The list (or order) of jobs may need to change, and unexpected conditions may prevent completing the installation in one visit. Success may need to come in stages, especially at remote and less-developed sites. Your first objective may be to establish a secure enclosure to store supplies and support later visits. Getting materials to a site can be more difficult than getting yourself to the site, so watch the weather and try to get structures closed-in, secure, and supplies on-site before access becomes more difficult.

Workmanship issues become important when reliability is concerned. Take the time to do things methodically and right. Check things out as you go, so that you can to rely upon whatever has been put in place. Don't assume you'll get back to correct some detail later; do it right the first time. Expect things to take longer than you want and you will rarely be surprised. Take a flashlight, because the day may end before you are done with the job. The penalty for oversights can be a flakey or inoperative installation, and needing to re-visit the site to correct what should have been done right the first time.

While doing the installation, note any unexpected situations. Wind whipping cables on a tower may lead to broken wires: Do you need to lash them down more securely, or route them in conduit? It has been said that most battle plans do not survive first contact with the enemy, and the same may be true for new designs. You may need to alter the plans during the installation. This also suggests that taking more than a minimal kit of tools and supplies may be useful to handle simple design changes.

If follow-up visits are required the checklists will be handy to make sure you take the right supplies and tools for the job. Keep adding to the lists as emergent issues and needs are identified, either in the field or as you prepare for the next visit. It's annoying to forget a critical part, and it may be a long time until the next visit.

If you expect to maintain the hardware in the future, doing (or at least being present during) the installation will help assure that you know where things are located. This can make subsequent visits less traumatic when you have to go it alone. Consider taking pictures of the installation, to help remind you of otherwise undocumented details. Pictures can also help clarify details when enlisting the help of others.


Site Operation

Hopefully your site will have a long and trouble-free life. More likely, time will reveal some problems.

As mentioned in the "Site Design" section, trying to anticipate and defend against all possibilities is pretty difficult. We never have enough time, money, or wisdom to avoid every possible problem. By monitoring the system's performance and being observant, you can use time to help identify the shortcomings which are real and worth fixing.

Consider each problem as an opportunity to improve the design. They represent oversights for situations you failed to anticipate, or chose to handle, or could afford to handle adequately at the time. Continuous improvement can significantly and quickly improve a site's performance and reliability.

You might consider enlisting the aid of others. They may see problems that you don't, and have improvement ideas worth considering.


Network Management

We had five camera sites which are linked together on a network using wireless "bridges". The hardware is very flexible, which is another way of saying that it takes a bunch of setting configurations to get it running the way you want. We encountered the following considerations:

We chose to use "static" IP addresses, which are determined by the network's system administrator. (That could be you.) Addresses are assigned once and do not change. (DHCP is "off")

Every "bridge" box, network camera, video encoder and computer on the video network has its own unique IP address (e.g. "192.168.1.241"). Addresses are specific to this network and it's best if you keep a written record of the assignments. You'll need to know what IP address to assign when the next piece of equipment (like a visiting laptop) wants to join the network.

We created two types of local HTML web pages to work with the hardware on the network. The first page created was for network administrative use. It is used to test for connectivity to every IP address. Every live and static video channel on every video encoder and network camera can be tested from this page with a single click. The page's HTML looks like:

Filename:  index_admin.htm

<HTML>
<BODY>
<H2> Wireless Network Development Test Page </H2>
<P>
<HR>
<H3> Wireless Bridge boxes </H3>
<UL>
<LI> <A HREF="http://192.168.1.230"> 192.168.1.230 - At South Lodge, to Easy Rider  </A> (login / password)
<LI> <A HREF="http://192.168.1.231"> 192.168.1.231 - At Easy Rider,  to South Lodge </A> (login / password)
...
</UL>

<HR>
<H3> Nodes on the wireless network </H3>

<P>
<UL>
<LI> <A HREF="http://192.168.1.239"> 192.168.1.239 - Base area temperature sensor </A> &
     <A HREF="http://192.168.1.239/setup.html"> set-up page  </A> (login / password)
<P>
<LI> 192.168.1.241 - Lodge -
     <A HREF="http://192.168.1.241"> 4 channel video encoder </A> (login / password)
<LI> 192.168.1.241 - Lodge - camera #1 -
     <A HREF="http://192.168.1.241/index.html?cam1=on&size=1&columns=2&mode=0"> live </A>
   - <A HREF="http://192.168.1.241/jpeg.cgi?0"> static </A>
...
<LI> 192.168.1.241 - Lodge - camera #4 -
     <A HREF="http://192.168.1.241/index.html?cam4=on&size=1&columns=2&mode=0"> live </A>
   - <A HREF="http://192.168.1.241/jpeg.cgi?3"> static </A>
...
</UL>

</BODY>
</HTML>
("..." sections represent many additional lines which look very similar.)

The second HTML page was created for local users to access the camera views. It relieves the users of needing to know anything about IP addresses or the URL syntax: All they need do is click upon the view which is of interest to them. A "main" page displays six static pictures, and updates itself every 6 seconds. Hyperlinks connect to other pages, of which there is a separate page for each group of cameras, plus a page for each individual camera. This collection of web pages can be copied and used by any computer connected to any port on the network.

Filename:  index.htm

<HTML>
<HEAD>
<TITLE> Hoodoo Butte pictures (6 second update) </TITLE>
<META HTTP-EQUIV="Refresh" CONTENT="6">
</HEAD>

<BODY>
<TABLE BORDER=5>
 <TR>
  <TD>
   <CENTER> 
    <A HREF="index_Grn0.htm"> <! Full-rate URL here>
     <IMG SRC="http://192.168.1.245/jpeg.cgi?0" WIDTH="320" HEIGHT="240">
     Green Chair top </A> &
    <A HREF="index_Grn.htm"> related cameras</A>
   </CENTER>
  </TD>

  <TD>
   <CENTER>
    <A HREF="index_Ed0.htm">
     <IMG SRC="http://192.168.1.243/jpeg.cgi?0" WIDTH="320" HEIGHT="240">
     Ed Chair top </A> &
    <A HREF="index_Ed.htm"> related cameras</A>
   </CENTER>
  </TD>

  <TD>
   <CENTER>
    <A HREF="index_Base.htm">
     <IMG SRC="http://192.168.1.241/jpeg.cgi?1" WIDTH="320" HEIGHT="240">
     Base area </A> 
   </CENTER>
  </TD>
 </TR>

 <TR>
  <TD>
    ...
  </TD>

  <TD>
   ...
  </TD>

  <TD>
   ...
 </TR>
</TABLE>

<P>
<UL>
<LI> <FONT COLOR="FF0080"> Click the "<A HREF="index.htm">REFRESH</A>" button, or any picture,
     or any link to continue if this page stops updating.
     </FONT>
  <UL>
  <LI> The "failure to update" problem does not seem to happen if two computers are running this web page.
  </UL>
</UL>

</BODY>
</HTML>

The following HTML page displays displays a single camera's live picture, and returns to the main page after 60 seconds:

Filename:  index_Ed0.htm

<HTML>
<HEAD>
<TITLE> Ed Chair top (live) </TITLE>
<META HTTP-EQUIV="Refresh" CONTENT="60;url=index.htm">
</HEAD>

<BODY>
<iframe src="http://192.168.1.243/popup.html?0"
        width="725"
        height="500">
</iframe>

<CENTER>
 RETURN TO
  <A HREF="index.htm"> Main page</A> or
  <A HREF="index_Ed.htm"> Ed Chair camera group</A> now,
   or automatically return after 60 seconds
</CENTER>

</BODY>
</HTML>

The "index_Ed.htm" page referenced in the above web page shows a live view from the primary camera at the site, plus static pictures from the other cameras in its group. Clicking upon a static picture allows the user to see live video from that camera. Limiting these web pages to show one live picture at a time is intended to minimize the network traffic.

Before a new box goes on the network it needs to be configured, where at a minimum its network address is set. The network's system administrator determines what the IP address will be. A CAT5 "cross-over" cable may be needed to directly connect between the configuring computer and the new hardware in case the device's default IP address is incompatible with the network.

Hardware can be operated anywhere on the network where it can be plugged into an available port on a "switch". New cameras can be tested and configured in the office, before being deployed in the field. A laptop computer can visit each site, yet can check-out equipment located anywhere on the network. This transparency removes the need to go to a specific location to see how something is working.


Configuring the Hardware

Most of the hardware requires some amount of configuring to get it ready for operation. While there's no substitute for reading the instructions, here's an idea of what's involved.

Wireless Bridge

Determine the unit's default IP address, like "192.168.1.1", from the package or turn-on documentation. Connect the bridge to its power source, and to the network (or computer), and from an MSDOS window try to "ping" it (e.g. "ping 192.168.1.1"). You should get a 1 or 2 msec response time and 0% lost packets.

Access the box's network interface using an internet browser, using a URL that contains the IP address and looks like "http://192.168.1.1". You should look at and probably change many settings:

5-port Switch

The easiest box to set-up:

Video Encoders & Network Cameras - Stardot version

The initial set-up uses a serial "null modem" cable which comes with each unit. Installation software is provided on a CD. If your computer lacks a 9-pin serial port you will need something like a USB-to-RS232 adapter. A separate computer with a serial port & CD reader could also be recruited.

Once the static IP address has been set, subsequent configuration work can be done using the unit's internet interface. The URL looks like "http://192.168.1.240", and a CAT5 crossover cable from the computer to camera makes the connection.

Power outlets

The "configuring" here is about being careful with how things are connected.

Having enough outlets for all the power plugs is one consideration. Keeping everything safe from "power surges" is a bit trickier, as the hazard can arrive over any external wired connection, not just the power line. The CAT5 cables make all network hardware at a node communally vulnerable. Each box, via its power cord, wall wart, and other wired connections, represent a hazard.

Power-wise, the wall wart for a powered speaker for a computer should be on the same surge protection as the rest of the node if the computer connects to the network with a wired (CAT5 cable) connection.

Signal-wise, a phone line going to a printer/fax machine connected to a computer on the network is also a risk. Without some form of surge protection it too endangers the network node.

Like the challenge of defending against "all hazards" in the design process, defending against damage from voltage spikes can either be done well to begin with, or you can figure it out after some hardware has been killed or damaged. Prevention should be cheaper than replacement, especially if it happens more than once.

Network Interface Card (NIC)

Prior to purchasing the NIC, a sales person asked about what operating system was used by the computer into which the NIC was going.

Visiting Laptop Computer or Netbook

These notes were almost unreadable, so guesses were made that should be similar to the proper spelling or word length ...

On a Windows XP computer:
- "Start", "Settings", "Control Panel", "Network Connections",
- "LAN or High-Speed Internet", "Local Area Connection [2]", "General" tab, "Properties"
- "Internet Protocol TCP/IP", "Properties"

The general idea is to get:


Final Thoughts

It would have been handy to have been an expert in this field before starting. Instead, this has been a learning experience. Much was learned by doing things, sometimes the wrong way. Other learnings came from the internet while shopping for information and products. Some experience came from watching the system misbehave, and then trying to prevent it from happening again. Some design philosophy was realized while writing this page. Hopefully you find learning from other people's mistakes cheaper, faster, and more entertaining than doing it yourself.

The hardware noted here was selected after shopping the 'net for product features and prices. No doubt there will be newer, better, and cheaper versions available in the future. Good luck on your hardware selections.

Some details still need to be corrected, but in general this page is good enough to give you an idea about what needs to be considered. A more detailed description is probably something I could not stand to write, and you would not want to read.

Feedback regarding this or related pages can be emailed to "".

2011 03/24