This also resulted in failure. When the rotational degrees are set to 0,90 the video rotates about its axis and when set to 60,120 there is a diagonal formation in the center yet the desired result is not achieved. The research to do this took a lot of my time and was still unable to figure out. Another problem is the ratio of the video. The video streams in a fixed ratio, which may result in problems while streaming on different devices.

Coming from a background with 0 programming skills it would just be a waste of time trying to figure out the code that I'm still unable to fully understand. Hence I had to think of alternatives.

Now instead of using pyramidal structure to project my hologram, the next alternative would be to use a single plane hologram

The code that I already have(Background subtraction) would work perfectly fine with this, the only addition needed is to stream the data from the cam to a mobile phone.

I tried googling on how to do streaming using python, and there most of the options pointed to using Django, Tornado or flask. I had help from a friend Jimesh to figure this out. We finally used Flask to stream data.

Flask:

Flask is a micro web framework written in Python. It is classified as a microframework because it does not require particular tools or libraries. It has no database abstraction layer, form validation, or any other components where pre-existing third-party libraries provide common functions.

I just didn't have the time to check out the complete documentation of Flask, as time was of the essence. Googling video streaming through flask resulted in some very good links that are were super useful in helping me use flask.

I pretty much used the same code from the above links and combined them with my background subtraction code. I first used a test code to check if the data is being sent and it worked perfectly fine.

To load the HTML page one needs to use the IP address of the network that the device is connected to and specify the port no. In this case, I used 8080 as my port.

Wirth the help of Jimesh I was able to complete my code and also add a button on the browser to make the video full screen.

Screenshot of phone and com.

And the same was opened on my phone browser as well.

Designing

I started sketching the basic form of the prototype that I wanted to build. There are several "Z" shaped devices in the market that works on the same principle. I didn't want to use that form for my design. I used multiple software to develop my project depending on my comfortability to use them. I started forming rough shapes on Sketchup. I wanted to know how the design is going to look like first and SketchUp being one of the easiest software to use, I used the same to develop my form.

I iterated the shape and size until I reached the desired form. The initial design was for a tablet but later I changed the design to suit a phone of a maximum screen size of 9", where most of the phones today are 5.5" - 6".

Once I was sure of the form I started drafting the panels in AutoCAD. Autocad is like pen tool for mem coping from Architecture background. It's easier for me to realize my design in 2D first and the best tool suited for that is AutoCAD.

Now that I have drafted all the sizes and panel I need to convert them to 3d again and assemble them to make press-fit notches. Its easier to do it this way as manual press fit notches may result in an error. I can use fusion and sketch up to convert them to 3d. I used Fusion as its accurate and exporting .dxf file is easier.

I assembled all the parts in Fusion and also rendered it to see how the end product may look like. I used black on all surfaces because the reflection of the image is seen better this way. Once I assembled all the parts I created notches in Fusion using extrude command. This cuts away the material resulting in notches and keys at the precise locations. I exported the dxfs and took it to Autocad again to give tolerance and nest it properly for cutting.

Laser cutting

The design was made for 5mm MDF, but unfortunately, only cheap MDF was available and the thickness was quite uneven. The average thickness of the material was 4.4 mm and I had provided a tolerance of .5mm after testing a small piece first. Using Power 65 and Speed 15 I cut my parts.

I assembled the parts, most of the parts fit in perfectly, but certain areas required slight hammering to fit in the pieces due to the uneven thickness of the material. The material was also bent adding to the difficulty in press-fitting it, but once assembled the for looked neat.

I tested the form by placing a wasted acrylic sheet as a screen and played videos to check the quality. The quality of the build was really fine and the projection was good, except that there was a double reflection. That might have been due to the thickness of the acrylic sheet. I had used a 4mm thick acrylic.

Apart from this, these are the design corrections that needed to be made:

-The viewing panel needs to be moved in order for the reflection to fall in the center.

-The Punctures needed to be provided to accommodate led strip lights

-the height needs to be slightly increased to facilitate 3d printed cover for camera s sensor

-patterns on the bottom panel for light

-area for circuit boards to be placed

-cutouts for the camera and ultrasonic sensors.

Once all the corrections were noted I took it back to Fusion, made the corrections in the assembly file and exported the dxfs to be cut.

I used the same parameters as before to cut the file but this time slightly reduced the tolerance.

Since I wanted the entire product in black I used Black (Matt)Acrylic spray paint to coat the MDF.

Note: Use spray paint with necessary safety precautions and in a well-ventilated environment.

The next task was to put all the parts together. There was slight difficulty in press-fitting again due to the varying thickness, but nothing that could not be fixed.

By doing so my final form was ready.

3D printing

Certain parts of the design required 3d printed components. The screen holder and the casing for the camera and ultrasonic sensor to name a few.

To design for the camera and ultrasonic sensor, I first measured the exact dimensions of the units and cadded them.

From here on things became very simple. I made an outer casing and an inner bracket to hold the camera with an opening on one side for the wire. The sensor design was inspired from one of the open source designs.

Also made an attachment to screw them to the front panel with 2.5mm screws. The dxf was then imported into fusion and extruded.

Took it to cura to slice and then printed it. The first print was unable to fit in the components properly. So I had to adjust the parameters and print again. And this time they all came along pretty well.

Spray painted the same to give it a uniform look.

The glass clamp was designed in auto cad first. Then extruded in Fusion.

The print came out quite nicely but was too small and was in extreme corners.

So I decided to modify the design and print it for a full length but in two parts.

Printing them was no difficulty.

I spray painted them and fixed them onto the device with 2.5mm screws.

This sums up all the design needed for my project.

Electronics

The basic concept of my electronic part of the design is, When the user comes in front of the device, an ultrasonic sensor(Input device) measures the distance of the user. When the condition is satisfied, LED strip light is switched on (Output device). So the components required are:

-Modular board (To control all the input and output functions)

-Ultrasonic sensor

-LED light

-Relay(to switch between Voltage for LED)

-Power board (To power up the device and to convert 12v to 5V)

Out of these, I have designed the modular board in week12 as a part of my output week. The process can be found here.

Now coming to the power board. I used LM1117 to convert 12V to 5V. I designed the circuit in eagle.

The drill bit that was used in the process is a very bad one. Unfortunately, we were not able to source any more bits. The milling was not all that great. It was quite rough. However, the boar was useable. This is the final board.

The pinout for the entire system is as follows.

I tested the circuit by connecting all the components. Everything worked really fine. The only thing that had to be calibrated was the distance sensor. To know what distance the led was suppose to glow. The fixed distance now is 65cms.

Integration

Now that all the systems are sorted, its time to put everything together and test it. The following picture depicts all the components used in the project

The first part is to attach the ultrasonic sensor and the camera.

The wires are routed on the sides and are taken at the back through a channel specially made to clear the wire clutters. This makes the packaging better.

I had made a sliding cover on top to conceal the product fully.

The led lights were stuck on the inner surface and on the bottom panel.

The bottom panel holds all the circuits and the power board. The wires are well concealed there.

On integration, I noticed that the led in the front was too bright. This disturbed the reflection on the acrylic. So I 3d printed a small cap with transparent PLA.

I spray painted it on one side to match the tone of the device and double coated the top with tape to disperse the light.

This is the comparison between the two lighting conditions.

The bottom panel also lit up quite nicely.

Now the final part is to test the device and its functionality. The following video explains the making and the working of the device.