22 Jan

Ticking Off the To-Dos

This weekend I was able to do a few tasks I always said I wanted to do, but never really could convince myself that I would find the time to do them. These projects involved cutting wires, fishing wires through walls and insulation, moving furniture, de-cluttering wires, and completely wiping my computer’s hard drive and doing a clean macOS install.

I’m impressed at least.

As you may recall the outdoor webcam was hardwired  to an (old) Airport Express via a POE injector.  The (old) Airport Express was then plugged into an outlet in the basement, and connected to the main wireless network. It was a hack. A temporary solution to something it took me 2 years to clean up.

Now the cable that comes from the outdoor webcam goes up through the floor of the basement right where it comes in the house. Then it comes out right behind the entertainment center in the living room where it connects to the POE injector and the network switch connected directly to the Airport Extreme. A much cleaner solution.

In order to get that done I cut a few dozen feet of old wire from when we had Dish Network, and even an old telephone cord. That knot of cabling was unsightly, and even more so with the entertainment cabinet now up on legs. Once they were gone I was able to fish a wire coat hanger up through the hole in the floor and up into the box in the wall. It took a few tries, but with a little help from my wife we were able to get it done.

Then this morning I did something I have been threatening to do for many months, but was always a little afraid to start. I completely reformatted the hard drive on the iMac, and reinstalled macOS. The threatening I mention was more planning then shaking a fist at the machine each time it failed at something.  I made backups. I wrote long detailed instructions on how the camera worked, and I took screenshots.  I copied important files to off-site cloud backups, and I made sure I had a few good Time Machine backups.  I used Notes to create a list of Must Have software, and I ticked off the list slowly over the day.

Right after installing Dropbox and 1Password, I got the webcam and all of its processes up and running.  Not too much down time, but today’s timelapse won’t be much, and I forgot to enable the daylight timelapse until it was dark, so that likely will be missing completely.  But everything should be back to normal by Tuesday, and so far everything is humming along.

Fingers crossed.

05 Jan

Options Are Always Nice

A few days ago I wrote about the new features I have added to the FarmCam. While writing the entry I stumbled upon actions settings in EvoCam that allowed me even more flexibility in the ways I share the views. However, I had a few concerns before I could decide if it would be a viable replacement for the current method.

My experiments in the last few days have resulted in mixed — but encouraging — results, and as promised I am sharing them with you.

Good Things I Learned

EvoCam export action settings

EvoCam export action settings

First, a few of things I like about EvoCam’s time lapse export feature:

  • The process is self-contained. EvoCam acts as the file converter and the FTP client. I already use the FTP settings in the software to upload the still images, so when the time lapse movie is created by EvoCam, it easily uploads it to this website without having to use a third party FTP client.
  • Control over the frames per second (FPS) settings.  With the current process, I have no control over the frames per second, meaning that the speed of the daily video is going to be 30fps.  With the EvoCam settings I can decide just how fast the individual frames are displayed in the finished product. I have been experimenting with frame rates as low as 5, and as high as 15. The results are pleasing.

Minor Downsides

Some downsides do exist, but EvoCam helps minimize some of the impacts:

  • By slowing the video down, I make for a longer watching experience. This wouldn’t be an issue if most of the videos — especially in winter! — weren’t complete darkness; however, since making the adjustments to the white balance settings, that is the case. The solution to this, so far, has been to use EvoCam’s light detection feature. Each video starts just around daybreak, and ends just after sunset. The drawback, is that any celestial movements such as the moon or bright stars are missed.

    EvoCam light detection settings

  • File size concerns. I would love to include the night time hours for the reasons above. Slowing down the time between images in the video means the video’s file size may get to the point where I lose a portion of my (already very small) audience.  The experiments with daytime-only videos are resulting in videos that are of a manageable size, but that is also a result of using stills taken 60 seconds apart as opposed to the 30 second intervals I currently use. More images, and slowing things down could mean a much more bloated final video. I have the bandwidth thanks to my awesome ISP (also my employer), but many of my neighbors either have slow DSL or services such as satellite with bandwidth caps.

Further Questions to be Answered

One the most frustrating thing about all this is that EvoCam’s developer disappeared without a trace over a year ago, and I worry about the future of my little hobby. I have been unable to find similar software that does half of the things that I want, nay, need for my projects. For the time being the software is stable, but I wish that someone would be willing and able to take it over if the former developer is done with it. If you know how this can be accomplished, please do hit me up.

For now, my experiments continue.  I still have a few things I need to investigate. For example, I am not sure how to control archiving or names of the files. I am fine if the local files are simply sequential (i.e., daily-1.mp4, daily-2.mp4) because I will have the creation dates to let me know the date of the footage. I like to review the videos to see if they contain anything funny, unusual, or unique.

But, I need the uploaded file to always be the same (daily.mp4) so that I don’t have to adjust the HTML. Based on the results when the “delete movie …” setting is enabled, I assume this will be the case; however, as usual, I just thought of this whilst I was typing the paragraph above.

Parting Thoughts

As I said in the opening of this entry, there are some benefits and drawbacks to both methods. The great news is I don’t really have to choose. The current method is hacked together, but it is working (knock wood). And I can continue to run my experiments without them getting in the way of the other method.

Right now it is looking promising that I can stop running the previous method, and do all of the things above within EvoCam. If I do, you’ll be the first to know.


UPDATE: 7 January, 2018

Found out that if one does not check “delete movie…” EvoCam saves only one instance of the resulting video, overwriting the previous day.  This is far from a setback; my quick solution was to create a three-step Automator workflow to rename the video file each day, and move the file to a local archive directory. However, if this is to be a viable replacement from my hacked method, I will need to figure out what time to stop the EvoCam video, and what time to run the cronjob for the archiving workflow. This isn’t an issue with my test model as I am only recording during daylight hours.  But if I want the video to run from as close to 00:00:00 to 23:59:59 as possible, then I need to time things properly. EvoCam needs to export the video to the local folder, upload the file via its built-in FTP. Then the cronjob needs to run an upload the resulting video before the new video is created.

Never mind, the above is all wrong! The way EvoCam works is it hold the images for the new day’s video until it is time to make the video.  Therefore, it I don’t need to time when I run the cronjob. I feel silly now.

01 Jan

Happy New Year?

I have been thinking a great deal lately about the toxicity of the so-called social networks. From the hate being spewed on Twitter, often from “bots” designed to create discord and spread misinformation, to world-wide manipulation of democratic elections.  It occupies my thoughts more than I would like, and until something dramatic happens either by government or the corporations themselves I don’t see more than one solution. Stop using the corporate sites, and look for alternatives.

Much like drug dealers, Twitter, Facebook (and others) give their stuff away. Why? Because they want us to stick around. We are the product. These dealers want us to share more information: add more tags; tag more people (facial recognition!); geo-tag pictures (conveniently embedded in digital photos!) from where we post content. Plus the things we like — or dislike, or love, or makes us sad or angry or hate or even laugh. Constantly.

That data is then mined to create profiles on us. Couple this with benignly-named cookies you pick up as you enjoy the (currently) open Internet, you let Facebook, Twitter, Amazon, and well, pretty much anyone who has a website, know exactly everywhere else on the web you have visited *.

Wow, I really do sound like a conspiracy nut. And you should see the stuff I deleted! But stick with me. It’s really not outrageous.

I too am a participant. I have been on Facebook and Twitter for 10 years now, and have shared countless photos, restaurant reviews, check-ins, etc. I have been using the Internet openly (for the most part) for more than a quarter century. My domain, and most of my handles are my real name. I am not trying to warn people off this stuff. No one listens to me anyway. I have no megaphone or amplifier.

And, honestly, I personally think it is too late at this point. Capitalism, governments, corporations, are all too big and intertwined at this point.

Ok, I do sound more wack than woke.

But it all feels very different now. Learning how people’s feelings and beliefs were manipulated by others in order to sow misinformation and hate, has left me with a sour taste in my mouth, and my opinion of humanity is at an all-time low.

The fact that Twitter will ban bots designed to help uncover and expose these horrible people, and not the hateful people themselves makes me angry.

Anyway, I have a couple of thousand words on how unsettling it all is. Writing this entry has been temporarily cathartic.

This is being posted at midnight of the new year, and it is the beginning of an experiment. Let’s see if anyone notices.

31 Dec

New Features for the FarmCam

In October my Australian cousin Peter, challenged me to create a different sort of time lapse.  He wanted to see how the view would look at the same time each day throughout the year.

I knew this would be easily accomplished with EvoCam by adding a new action set; so I set up a test, and let it run from 17 October through the last day of 2017. I used EvoCam to manually export the video to MP4 at 5 FPS. Below is the result.

During this experiment I learned a few things:

  • EvoCam software has location abilities, so I am able to have it do things at sunrise and sunset for my location. Therefore, in addition to the Daily Noon image, I have added a Daily Sunrise and Daily Sunset. These are still photos uploaded daily, and can be viewed with the other stills and videos on the FarmCam page. I will make time lapses from the images, but not sure if they will be seasonal or yearly.
  • EvoCam may be able to allow me to cut out my buggy Automator cron job. Right now the encoding from the EvoCam MOV to an M4V is done via default settings in Automator. Then I use the Automator app to convert the M4V to MP4 and upload the file via FTP.  I am running tests now to see if EvoCam can export the final video as an MP4 rather than a MOV. It has a feature for uploading the resulting video to the web server, which is a good thing.
  • If EvoCam can export as an MP4 then I can control the frames per second for playback as well. This allows me to slow down the final daily time lapse video, therefore making it more enjoyable viewing. Things that currently flash past, like animals, birds, cars, sheep, llamas, dogs, etc, will be visible.  The downside is that the video itself will be longer.  That may mean I will need to take stills once every 60 seconds instead of every 30 in order to reduce the final run time for the rustling time lapses.

I should have the answers to the experiment by tomorrow, and I can then make decisions.  Updates will follow.

Fingers crossed.

12 Dec

FarmCam Update – Dark Nights or Green Screens?

Example of Green Phantom effect.

For years I have been frustrated by a nagging issue where the screen of the webcam goes green.  Sometimes in flickers, and sometimes for long periods.  I always assumed the issue was something over which I had little control.  When I was using a digital camcorder, I had to hack it by leaving a tape out to have it in continuous record mode. Plus I was using an RCA to USB converter to get the signal from the camera to the iMac.  I just assumed there was something amiss with my hardware.

Example of Night Vision feature.

When the issue continued after I introduced the current outdoor, networked camera I still blamed my hack jobs.  The camera is currently connected to a POE which is then connected to an old Airport Express, which in turn is an extention of the wireless network to which the iMac is also connected. You can probably see why I would assume the issue is with the creator of this Rube Goldberg contraption.

So convinced that I never even tried to research other options.

However, lately the amount of green being captured and displayed has been a source of frustration, and I decided to take a stab at potential solutions. I had already considered using the extra length of the Ethernet cable from the camera, and running it directly to the Airport Extreme router, which would remove the POE and the Airport Express from the setup. Or even having the camera wired directly to my Apple Mac Mini server and run all the software on it rather than my office desktop.

Current night view … with LED lights on barn.

I still would like to do one of these things, but some poking about led me a random reply on a message board where the person suggested making “any adjustment” to the white balance settings. White balance in digital photography essentially refers to adjusting colors so that the image looks more natural. Most light sources do not emit purely white color. They have what’s referred to as a color temperature. For the most part I rarely have never given much thought to white balance beyond the default settings.

Logging into the web interface for the Hikvision camera, I looked at the settings all of which are presets. I tried each preset and sometimes the results where obvious and other times not so much. The obvious results were mostly the extremes.  The image would have a bright green shading or it was crystal clear. I also noticed I got different results if I adjusted the night settings.

My camera’s White Balance settings.

The plan is to try these new settings and watch the daily timelapses to look for evidence of the green screens. If none materialise then I will make note of them, and try tweaking the settings to see if I can improve the results or not.

The bad news is that I may have to forgo any night vision settings, which is disappointing as I just replaced the recently burned out lights on the barn with new LEDs.  The good news is that the daytime image has improved significantly, and so far the “Natural Light” white balance setting has not shown any sign of the green phantoms.

Stay tuned, and let me know if you ever see any unusual images, frequent green screens, etc. I’ll report findings later.