Photo Animation - Part 2

Paternal Great Grandparents Come Alive

My great grandmother, Scholastica Ostrokolowicz, died in January, 1921, twenty-seven years before my birth. Her earliest picture in my possession is from approximately 1888, about one year after her arrival in America from Poland. A colorized version of this picture is shown next. Scholastica sits with her bearded husband, my great grandfather, Jan Ostrokolowicz. Standing to the left is her young son, Xavier. Her approximately 15 year-old daughter, Ursula, stands behind. I believe my grandmother, Barbara, sits on Scholastica's lap.

I was able to animate Scholastica's 31 year-old face from the old, faded picture.

Unfortunately, I could not animate great grandfather Jan's face from the same picture. The animation tool would not recognize his beard and faded facial features. 

About 6 years years after the first picture above was taken, in approximately 1894, another family picture was taken which I was also able to colorize. Now Scholastica is about 37 years old, her daughter, my grandmother, Barbara, on the right, is about 7 years old, and Barbara's younger sister, Johanna, is on the left. Enhancement of the colorization put some erroneous purple and red tints on dresses, but made faces sharper.

Next are two animations of Scholastica's 37-year-old face. The first colorized animation had trouble keeping Scholastica's hat attached to her face. For some reason processing a black and white photo version resulted in more natural movement in the second animation.

Years pass before Scholastica's last picture was taken. Colorization gave life to an old black and white photo. I can only guess at the year. Her daughter, Johanna stands beside her as a young woman. Perhaps the date is between 1914 and 1921. If so, Scholastica is between 57 and 64 years old.

Animation of the black and white photo worked better than the color animation. I was able to chose a version which included a small smile to soften Scholastica's otherwise stern expression.

Enhancement worked miracles on this portrait of elderly, dignified Scholastica which hangs in our bedroom.

Only two old photos of my great grandfather, Jan Ostrokolowicz, produced reasonable animations. The most interesting picture, from about 1892, shows Jan in his Polish cavalry uniform standing with his son, Xavier, on the right. Jan is about 45 years old and sporting a huge beard. The original photo was damaged, making it impossible to repair Jan's missing foot and rifle bottom. 

The animation tool frequently has trouble keeping hats and beards attached to faces. In this case Jan's helmet followed his head, but the beard movement is a bit odd.

The next picture is from about 1898. Jan is approximately 51 years old and has lost his beard. I colorized the original black and white picture, but didn't enhance it. This prevented erroneous purple patches on dark clothing. My grandmother, Barbara, now about 12 years old, stands on the left.

Below are two animated versions of Jan's face with slightly different movement in each. I chose these because they exhibited the least hat distortion. They both show brief hints of smile which really bring Jan to life. 

 

Jan Ostrokolowicz is like a ghost. I've searched and searched for any documented record of him and failed to find any beyond his name on a passenger list from the ship that brought him to America from Poland. I know almost nothing about him.

I continue to be fascinated by these animations!  


 

 

 
 
 
 

Photo Animation

Grandfather Comes Alive!

I recently discovered a photo animation tool available on MyHeritage.com called "Deep Nostalgia". This remarkable application can produce limited lifelike movement from still photographs.

I never met my paternal grandfather, Michael Michalik, who died at age 45 in 1927. Only five pictures of Michael remain in my possession. Two photos show him as a young man about 20 years old. Three other pictures show him looking worn out shortly before his death. I used the first image below to bring my grandfather back to life as a young man. It's a poor quality black and white picture from somewhere between 1901 and 1905. Michael stands behind his brother, Ignacy, and Ignacy's new bride, Caroline.

In the first step, I cropped the image to show Michael alone.

Next, I adjusted levels in Photoshop to darken the haziness.

Then I enhanced the previous image using the MyHeritage automatic enhancement tool. The amazing enhanced result clears away many fuzzy imperfections of the original picture.

Then I colorized the previous enhanced image using the MyHeritage automatic colorization tool. Overexposure of Michaels's forehead is more obvious in color than in black and white. His dark jacket also has inaccurate purple areas. This seems to be a common flaw in many colorizations of dark clothing.

Finally, the spectacular "Deep Nostalgia" animation tool brought Michael to life as you can see in the following short video.

The animation works incredibly well! Apparently, Michael might have been wearing a hat before the photo was taken because his hair is crushed in along the left side.

Animations work best with isolated, forward facing heads with no beards or hats. It also works best if smiles are subtle without too many teeth showing. Animations often add brief, small, charming smiles to originally unsmiling faces. For example, check out the hint of a smile near the end of a different less enhanced animation of young Michael.

Another photo from near 1900 shows Michael standing to the right of someone named Karoll Lichtman. I wish I knew the circumstances of the picture. The colorized version looks particularly good. There are no purple patches on the dark suits.

Michael's animated face from this picture seems more rounded than the previous animations. His hair is also slicked down.

A slightly different animation was produced from the black and white original version. I don't know why "Deep Nostalgia" treats different levels of enhancement differently, but I like the nice little hint of a smile at the end of the uncolored animation.

The next 25 years or so after 1900 were not kind to Michael's face. In the next picture he stands between his wife, Barbara, to the right, and Barbara's niece, Victoria, on the left, in the year 1926.

The colorized animation of Michael's 1926 face did not work as well as the black and white one shown next.

Although these videos are short and artificially produced, they brought my unknown grandfather back to life for a few wonderful seconds.
 

 
 

 

Backyard Light Pollution

Too Many Lights!

When we first moved to our new home in Williamsburg I was thrilled with prospects for backyard astronomical observing. No large trees blocked the view. Neighboring lots were initially empty. There was an unobstructed view of the southwest horizon. It wasn't long, however, before vacant lots were occupied by two-story houses blocking nice horizon views. Then neighbors moved in. Now my backyard is surrounded by houses, each equipped with rear spotlights which make nighttime observing nearly impossible.

Imagine standing on my backyard observing pad with telescope ready to go. Facing southwest you see this view.

Turning slightly right to look more directly south reveals another glaring light seen below.

If I'm unlucky and both neighbors light the night at the same time, I see this.

Now turn about 90 degrees farther right to face west. Here are the most troublesome discouraging lights of all.

The upper beacon in the previous image is an extremely bright interior light, not an outside spotlight. It blasts out through a semicircular second story window lacking blinds or curtains. It shines directly in my face whenever I step into the yard to check the sky, and it rarely turns off. This nemesis light often remains lit well past midnight, sometimes all night long! Underneath the awful second story light is another ever shining first story interior light glaring through open blinds. I can imagine humbly asking other neighbors to please turn off their outdoor spotlights, but how can I ask someone to turn off lights inside their home? Maybe a bold character could do this, but I'm not that person!

Now turn right another 90 degrees towards southwest to find the remaining awful lights.

To be honest, all five nightmare lights seldom shine simultaneously. In fact, the two northern lights only come on during infrequent nighttime barbecues. One of the southern lights usually shines temporarily to let a dog out. The other southern light, however, is frequently on continuously in the summer. Still, it actually takes only one active light to ruin observing.

 

The worst lights are the interior house lights in the east/southeast. They glare out all night almost every night year round. By unfortunate alignment they shine directly on my observing pad. If I move to the edges of my yard, window frames block direct sight lines to these awful lights. Unfortunately, telescopes can't be set up at these places.

 

What's the probability of finding all five lights simultaneously off? Very low! Sometimes I dream about a power outage. Power outages from storms aren't often accompanied by clear skies, however. 

 

From time to time, fool that I am, I set up a telescope for observing on some clear moonless night. It doesn't take long for frustration to set in as, seemingly at random, spotlights turn on and off destroying whatever modest dark adaptation I manage to accumulate. It's very discouraging. Then I'm reminded why I chose to specialize in daytime solar imaging rather than nighttime deep sky imaging.

 

I've fantasized about ways to block the lights, but possibilities all seem impractical or too expensive. My favorite fantasy is motorized black fence panels that elevate all around my yard blocking each and every light! Of course, I could travel to a darker location, but this is increasingly unattractive the older I get. When I was younger I would spend all night driving around the countryside setting up a telescope at various places. I could do this for hours past midnight, then get up the next morning for work. At this point, however, I can't summon the energy required to pack up equipment, haul it to a dark site, set it up, face possible deteriorating sky conditions, break down equipment, store it back in a vehicle, haul it back home and unpack it all. The convenience of backyard observing is too great in comparison. Traveling is just impractical. So, I'm stuck with backyard light pollution.

Solar Energy Production - Part 2

Seasonal Changes

My solar panels have been operating for six months, long enough to go from the June summer solstice to the December winter solstice. The panels have now experienced a full range of seasonal solar altitude and azimuth changes from maximum to minimum. As expected, energy production steadily diminished from June thru December as daylight grew shorter and solar altitude decreased.

The panels became operational on June 19th, one day before the summer solstice. Clouds, of course, affect power output, so I found the clearest possible cloud-free days to track seasonal changes. The first nearly cloud-free day after June 19th was June 26th when the panels produced their biggest amount so far, 20.63 kwhrs as shown in the following image. Some thin clouds appeared in late afternoon on June 26th, so I'm sure the absolute summer solstice maximum is slightly greater than 20.63 kwhrs.

The autumnal equinox happened on the morning of September 22nd. This day was, conveniently, completely cloud-free from sunrise to sunset! Energy production on September 22nd was 17.25 kwhrs as shown below.

The winter solstice happened on the morning of December 21st. Unfortunately, December 21st was completely overcast. December 26th, five days later, was the completely clear day closest to the winter solstice. Energy production on December 26th was 11.81 kwhrs as shown in the next image.

When no clouds interfere, energy production curves are nearly symmetric about their peaks except for small diminishing tails on the right side just before sunset. You can see the curves narrow as daylight hours decrease. Energy production is greatest while the Sun is near meridian transit around noon. At the summer solstice there is a broad flat maximum peak value at 0.60 kwhrs. At the autumnal equinox peak production is still at 0.60 kwhrs, but the flat peak has narrowed. At the winter solstice there is no flat peak, and the peak maximum value has dropped to only 0.50 kwhrs.

I've given examples of the best possible days for solar energy production. What about the worst days? The worst day so far happened on November 12th, a dark, completely overcast day with rain. In spite of heavy black clouds, dim light generated 0.56 kwhrs for the day as shown below.

I can imagine even lower energy production on a cold, dim, snowy day near the winter solstice when snow covers the panels from dawn to dusk.

Based on solar panel performance during the past six months I should have installed fourteen panels instead of ten. Fourteen panels would generate close to my usual yearly energy use without exceeding actual energy use. I'll be investigating what's involved in adding four more panels soon.

Jupiter-Saturn Conjunction

Hurried Observations

An unusually close conjunction of Jupiter and Saturn happened after sunset on December 21st. In weeks leading up to this event both planets were moving west to east relative to the stars. Jupiter, nearer the Sun, moved eastward faster than Saturn. So Jupiter steadily caught up to Saturn, like a faster car moving to overtake a slower one from behind. Four days before minimum separation, on December 17th, I rushed outside and looked for the planets in the southwest after sunset. Clouds and overcast were forecast after the 17th, so I wanted to take advantage of clear skies. At 6:17 pm EST I saw the following view captured with a two second exposure on my tripod-mounted Samsung phone camera.

The overexposed crescent Moon is about a half degree, or 30 arc minutes in angular diameter. Compare the Moon's diameter with the separation of the planets which was about 26 arc minutes at the time. Jupiter was brighter than Saturn and closer to the tree tops. The picture above roughly approximates the naked eye view. I spent several minutes moving the camera around to block streetlights and Christmas lights on nearby houses.

On December 21st skies were overcast and cloudy all day. It seemed unlikely to clear for the close conjunction. Nevertheless, I popped outside at 5:50 pm to check the sky. What an unexpected surprise to find clouds dissolving in the southwest! Hastily grabbing a coat and camera, I rushed outside to take the following picture at 5:59.

At first glance Jupiter and Saturn, closest to tree tops, look like a single star, almost indistinguishable as separate objects. If you click on the image to view it at full size, you can barely see dimmer Saturn as a bump on the right side of bright Jupiter's disc. Three stars lined up above the planets are in the constellation, Capricornus.

I increased camera magnification slightly to zoom in a bit and moved the camera to block bright lights. You can see the Saturn "bump" more clearly in the next image, especially if you click on the image to view it in full size.

A bit more magnification produced the next image.

Finally, I zoomed in quite a bit to the point where image quality was significantly worse. This showed the separation better.

Jupiter and Saturn were only about 6.25 arc minutes apart here! They would easily have fit within the same field of a planetary imaging camera mounted on my 130mm refractor, but there wasn't enough time to set up the telescope with all the necessary equipment. Instead, I hurried inside to haul out by big 25X100 binoculars. The two planets were beautiful companions in the same binocular field of view! Jupiter's moons were visible there also.

 

Saturn and Jupiter regularly pass each other about once every 20 years, but they haven't passed this closely and favorably placed for viewing since the year 1226. Do they ever get close enough for Jupiter's disc to overlap, or occult, Saturn's disc? Yes, but the last one happened 8,000 years ago. During this 2020 conjunction Jupiter and Saturn were almost exactly on the same line of sight from Earth. Since they were also somewhat close to the Sun (which had set only about 70 minutes before), Earth, Sun, Jupiter, and Saturn, in that order, were all almost lined up together along a straight line!

Although it was partly cloudy most of the next day on December 22nd, the planets were still remarkably close together. By the time I saw clouds clearing it was, once again, too late to set up a telescope. I thought the planets were too low near the horizon to see from my backyard where neighboring houses would block the view, but I discovered the planets were actually visible between two houses. I took the following image at 5:51 pm.

The foreground fence and houses were illuminated by bright spotlights glaring from surrounding neighbors. Separation on December 22nd was about 10 arc minutes, still astoundingly close, less than half a full Moon's diameter! Compare this final image with the very first image above from December 17th. You can see how Jupiter moved from below Saturn on the 17th to slightly above Saturn on the 22nd.

Sunspot Parade

Brief Observing Window

Skies cleared unexpectedly on November 24th, so I rushed outside to set up my solar telescope. There was a slight breeze on this cool morning where temperature ranged from 44 to 48 degrees. I used a ZWO 174 monochrome camera and 3X Barlow together with a powered USB hub to connect the camera to my laptop.

The following 4-image mosaic shows three major southern hemisphere sunspots that motivated my observing effort. (Click on the image for a  full-sized view.)

As the Sun rotates features move from left to right (east to west). On the right is sunspot 2783. Since my last observing session on November 21st, spot 2783 moved across the disc from left of center to right of center. A new pair of sunspots also rotated into view on the left. Smaller spot 2785 preceded larger spot 2786.

The following image of sunspots 2786 and 2785 is nearly 3-dimensional, especially when viewed in the enlarged version obtained by clicking on the image. In particular, look at the region just to the lower left of large spot 2786. The solar surface looks like waves in a shag carpet. There's a white energetic eruption emerging from the bottom of the umbra and another white eruptive plume popping up near the rim on the spot's left. Unfortunately, I didn't have time to move the field of view lower to fully capture the prominence on the lower edge of the image.

Sunspot 2783 was still accompanied by a modest white active area to its left.

After capturing the prominent sunspots I looked at the sky to check if it remained cloud free. Bad luck! Thin, hazy clouds and several contrails were increasing. I had to stop capturing images only 13 minutes after starting!

Modest Sunspots

Mild November Day

On November 21st temperature ranged from 57 to 64 degrees during a fine, cloudless, windless morning while I observed activity on the Sun's face. It's hard to believe I wore just a t-shirt in November! Two modest sunspots were visible, as you can see in the following imperfect 16-image mosaic. (Click on the image for a larger view.)

Sunspot 2783 had rotated to almost center in the Sun's southern hemisphere where it was accompanied by a white active region to its left and a filament to its right. Tiny sunspot 2784 was inside the small white active area on the left side of the northern hemisphere.

Constructing this mosaic was difficult. For some reason the photomerge package included with a new modern version of Photoshop Elements would not assemble the 16 component panels properly. A much older version of Photoshop Elements on a different computer was able to put the 16 panels together correctly into a circular disc, but it wouldn't blend their slightly different brightnesses. I finally tried an older program called, Autostich, which successfully completed the task.

I began the session with a ZWO 1600 monochrome camera, but the camera control software, FireCapture, wouldn't connect with the 1600 camera. Two weeks ago the 1600 camera worked fine with the identical computer and equipment. Two weeks ago all video downloads from camera to computer were fast and efficient. This week downloads were terribly slow. I have no idea why identical equipment in an identical configuration would behave so differently. All images here were made with a ZWO 174 monochrome camera. Thank goodness the 174 camera works reliably.

Next is a close view of sunspot 2783. It is a stack of 100 best frames from a 1,000-frame video.

The next image below, taken 59 minutes after the previous image, shows some changes in the region. In the enlarged picture you can see three new radial linear features in the sunspot's penumbral region. They look like mini-filaments above the dark umbra. The shape of the dark filament to the sunspot's right also changed. (Click on images to enlarge them and see details more clearly.)

Finally, tiny sunspot 2784 in the image below shows very small dark umbras starting to form. These eventually faded away in the following days.

Tiny sunspot 2784 was in the northern hemisphere, but recent significant sunspots have appeared mostly in the southern hemisphere.