- 1
- 2 632
Manuel Ángel
เข้าร่วมเมื่อ 25 ก.ค. 2020
DSuper8 project, based on the Raspberry Pi Film Capture project. Telecine machine to digitize old Super8 films frame by frame, using the Raspberry Pi.
DIY frame-to-frame telecine machine using Raspberry Pi 3 in operation
Telecine machine to digitize old Super 8 films using the frame-by-frame procedure.
The Raspberry Pi 3, which acts as a server for the commands of the main computer, is used to take images and control the movement of the film.
Once an image has been captured, it is sent via LAN to the main computer, which post-processes the image and saves it on the hard disk.
When the client program receives an image, it requests the following image from the server program.
Both the client and server program software have been written in Python 3.
A film stepping motor is used to move the film, which advances frame by frame with great precision, without the need to use a detector to take the image.
The operation may seem somewhat slow, but it is necessary to take into account that three photos are taken for each frame, in order to obtain HDR images.
The Raspberry Pi 3, which acts as a server for the commands of the main computer, is used to take images and control the movement of the film.
Once an image has been captured, it is sent via LAN to the main computer, which post-processes the image and saves it on the hard disk.
When the client program receives an image, it requests the following image from the server program.
Both the client and server program software have been written in Python 3.
A film stepping motor is used to move the film, which advances frame by frame with great precision, without the need to use a detector to take the image.
The operation may seem somewhat slow, but it is necessary to take into account that three photos are taken for each frame, in order to obtain HDR images.
มุมมอง: 2 632
Hi Manuel! Very interesting solution... I had the same idea and the projector is the same! But I don’t know anything about python...:/
I encourage you to build the telecine machine. You have to be patient, I have invested many hours both in the mechanical modification of the projector and in the development of the software. Some components such as the stepper motor and the lens I have ordered from a Chinese supplier and it took several weeks to arrive. Finally I have been very satisfied with the results obtained. After uploading the video, I have made quite a few improvements to the software, adding new functions and optimizing the processes. As for the software, if you liked my program, I could send it to you in some way. The GUI is in Spanish, although if you are interested, I could make an English version, or you can translate it yourself. However, the Python language is not difficult to learn. The GUI is made with the PyQt5 library, which you also have to learn to use. In the group there is also yart software, which could also interest you. Greetings and thanks for your interest.
Manuel Ángel thanks for answer! Yes, patient is very important! I started the project during lockdown days and now the RPI HQ camera arrives to finish the scanner! Do you have any social media to contact you? I have several questions... rsrs
@@ivansalomao5492 Hello again I do not have an account on any social media. I have a Gmail account and I am a member of the group groups.google.com/g/rpi-film-capture. I don't know if you've ever seen any conversation from this group. I suggest you become a member and we can communicate through the group or privately by email.
@@manuelangel8423 Nice Manuel! I don´t know about this group! I will follow now and keep in thouch! Tks again!
Hey there, very cool! Is that the OV5647 camera? What kind of lens solution did you have to focus on the film? Aside from the 3 images per frame, what is the limitation for speeding up the transfer, processor speed? Motor? Thanks!
Hi Wheaticus. First of all thank you for your interest. Yes, the camera is the OV5647. After multiple tests and trials, I decided to use it for the CS type mount that allows a wide variety of lenses to be used. The lens used is that of the link: www.google.com/url?q=https%3A%2F%2Fes.aliexpress.com%2Fitem%2F4000780942340.html%3Fspm%3Da2g0o.productlist.0.0.686b3054HVf4mf%26algo_pvid%3D077dbafd-e475-4d72-a595-629c94965eca%26algo_expid%3D077dbafd-e475-4d72-a595-629c94965eca-2%26btsid%3D0ab6f83915905129002514439e1113%26ws_ab_test%3Dsearchweb0_0%2Csearchweb201602_%2Csearchweb201603_&sa=D&sntz=1&usg=AFQjCNFdCc91KsFLXapkolcSaOuSnKYXiQ It is an economic lens and in my opinion it works well for this application. There are several factors that affect speed. The Raspberry Pi takes 3 images for each frame to be transmitted to the main computer via LAN. The stepping motor can rotate at a significantly higher speed, although with much rougher and noisier operation and many vibrations that would affect the stability of the lens and the camera. In my case the motor is 200 steps per revolution, but it is configured at the maximum resolution of 32 microsteps per step. In total, 3,200 microsteps are required to advance one frame, although operation is very smooth and precise. Finally, the main computer must perform several post-processing operations once the images sent by the Raspberry Pi are received, the most important of which is to apply the Devebec algorithm to the images, to obtain an HDR image from the received images. I leave you a link to the group where you can see construction details in addition to many other contributions. groups.google.com/forum/?oldui=1#!topic/rpi-film-capture/Ru-DZshqe_E A greeting. Manuel Angel Note: The original is written in Spanish and translated with Google Translator.
@@manuelangel8423 thank you for that information.