Category Archives: Learn and make

Continuous image capture from webcam

As part of an online course, I currently take “Computer Vision with Embedded Machine Learning” where is need to collect images to make the dataset. I modified the original code so I can use my regular Linux box and webcam, not Raspberry Pi. The programme counts down and saves images every 7 seconds. That isContinue reading “Continuous image capture from webcam”

Spurkfun Edge and Camera

My Arduino settings and code to use Sparkfun Edge with the camera module. Hardware Sparkfun Edge (Input only 3.3V.) Himax CMOS Imaging Camera – HM01B0 Wemos CH340G USB to serial (Make sure it output is 3.3V!) In Arduino IDE Add Boards Manager URL: https://raw.githubusercontent.com/sparkfun/Arduino_Boards/master/IDE_Board_Manager/package_sparkfun_index.json Install SparkFun Apollo3 Boards package Install SparkFun Himax HM01B0 Cameras library Settings Upload sketch PressContinue reading “Spurkfun Edge and Camera”

3. Sügavõpe: andmestikud, kadu, õpisamm ja mudeli parandamine

Minu Colab märkmikk. Fastai originaal Goodle Colab märkmikk. Andmestiku uurimine Andmestiku The Oxford-IIIT Pet Dataset veebileht. Kuidas on andmestik (datset) organiseeritud? Et näha, mis andmestikus on: Andmestik koosneb kahes kataloogist: images ja annotations. Veebileht ütleb, et märkuste (annotations) kataloog sisaldab, infot kus lemmikloomad asuvad, mitte seda mis nad. Kui meid asukoht ei huvita siis võimeContinue reading “3. Sügavõpe: andmestikud, kadu, õpisamm ja mudeli parandamine”

2. MNIST ja sügavõpe

Minu Google Colab märkmik aga parem vaata fastai 04_mnist_basic. Üks tuntumaid andmestike (dataset) on MNIST, mis sisaldab käsitsikirjutatud numbreid. Seda kasutati ühes esimeses praktilises käsitsikirjutatud numbrijärjestuste äratundmise süsteemis Lenet-5 aastal 1998. Laemealla MNIST_SAMPLE andmestiku. See sisaldab ainult numbreid 3 ja 7! Kuvame selle kataloogi (path) sisu: Masinõppe andmestikud (datasets) on tavaliselt jaotatud eraldi kataloogideks: treeningandmestikContinue reading “2. MNIST ja sügavõpe”

Andmekogu loomine jmd-imagescraper abil

Sügavõppe mudelite treenimiseks on vaja palju sildistatud pilte. Üks lihtsamaid mooduseid sellise andmekogu loomiseks on kasutada mõnd otsingumootorit. Otsingumootoriks on DuckDuckGo, mis kasutab Bing otsingumootori koostatud indeksit. Nii, et tulemused on üsna sarnased. Negatiivne külg on ,et leitud vasted on kõik ameerika kesksed ja ei arvesta lokaalsete eripäradega. Näiteks on siga mingi teibi bränd jaContinue reading “Andmekogu loomine jmd-imagescraper abil”

1. Sügavõpe (Deep leartning)

Lühi konspekt originaal fastai märkmikust 02_production.ipynb Masinnägemine Computer vision object recognition – asjade äratundmine object detection – asjade tuvastamine: asukoht, nimi segmentation – iga piksel kategoriseeritakse selle järgi, mis objekti osa ta on. out-of-domain data – Sügavõppe mudelid ei ole üldiselt head nende piltide äratundmisel, mis erinevad oluliselt struktuuri või stiili poolest nendest piltidest millega tedaContinue reading “1. Sügavõpe (Deep leartning)”

0. Sügavõpe (Deep learning)

Märkmed fast.ai kursusest enda jaoks. Ei ole ülevaatlik ega põhjalik kõikides osades. Originaal märkmik koos inglisekeelsete põhjalike selgitavate tekstiga: 01.intro.ipynb Vajalik on NVIDIA graafikakaart (GPU – Graphics Processing Unit). Need, mis on head mängimiseks ja 3D jaoks sobivad üldiselt ka sügavõppe jaoks. Google Colabis tuleb GPU kasutamine eraldi aktiveerida. Keskkonnaks Jupyter notebook. Kas oma arvutisContinue reading “0. Sügavõpe (Deep learning)”

Malevich

This artwork is inspired by famous suprematist artist Kazimir Malevich (1879-1935) and his painting “Black Square”. But I am not so radical. I have four LED squares inside one black square. Slowly blinking their different coloured lights. I need to make another video when there is not so light and you can actually see theContinue reading “Malevich”

The artistic shape detection algorithm

Today I learned one simple shape detection algorithm. In contrast image, it tries to find how many corners on some shapes are. When there is three, then it is a triangle and so on. And draws coloured contour around it. In a controlled environment, it works mostly as supposed. When you released it to theContinue reading “The artistic shape detection algorithm”

Motion detection on the webcam

It is surprisingly easy to make a small Python script that takes a webcam or any other video and detects when something is moving there. It uses the OpenCV library. 1. Difference between frames Compares two frames and displays only what are change. The rest is black. 2. Binary image Turn it into binary: onlyContinue reading “Motion detection on the webcam”

Custom Wooden Mechanical Keyboard

I had one old and cheap rubber dome keyboard with missing keys. So I take it apart and found this controller PCB board. The board pads were coated with graphite. I removed it and soldered wires on it. And put it on a breadboard. Then I mapped all rows and column combinations to find whatContinue reading “Custom Wooden Mechanical Keyboard”

How i made my digital radio

This is my simple one-button radio. One button to turn it on and change the volume. There are actually two buttons more: first to select a new channel and second one to save it to memory. It was a project that taught me how to draw PCBs, what are Gerber files and so on. IContinue reading “How i made my digital radio”