DEV Community: Inezayimana Henriette

computer vision

Inezayimana Henriette — Wed, 14 Oct 2026 09:26:53 +0000

i learn how machines learn to see: a beginners guide to computer version
so from pixels to predictions, and why its matters for africa
show your patterner a photo of a dog have duck on the head and he/she tell you what it is in a split second. they ever suggest if it is cute puppy.show the same photo to a computer and it sees something very different: a hugr grid of numbers.
computer vision is the field of al that teaches a computers to turn those numbers into a meaning. in this post i walk through how it works, from a single pixel to models that can find buildings from space, using the experiments i ran in hands on notebook
what computer vision ?
for me is to the computers get a meaning from images and video.
almost every vision system answer one or more of three qwestion : what is any image if itis cat,a cassavaleaf,a bus.
where it it? which pixels, which part of the picture
what happening? person crossing the road, a crowd forming( especially in video)
how computer sees a numbers: a digital image is a grid of tiny sqweres called pixels.
where each pixel holds three numbers: how much red,green,and blue it contains,each from 0 to 255. so a small 400*600*3=720,000 numbers. a modern phone photo has more than 30 million. the computer never see "a dog". it only sees these numbers of a computer verision mis going from numbers to meaning.
why 255? each value is stored in 8 bits, and 8 bits can hold 2 in 8 bits =255 different values, from 0 to 255. zero means none of that colour and 255 means full strength
my work i tested the photo of a dog in a babble bath, weearing glasses with a yellow robber duck on its head. i asked python to find the 10*10 patch of pixel with the most contrast, and it landed on the duck. one pixel there is full red, full green and almost no blue, and red plus green at full strength is exactly yellow .
the computer had no idea it looking at a duck ,only at a very yellow set of numbers. a color image is rearly three grid stacked together one each for red, green and blue. in the duck blue grid the values are tiny becouse a yellow object contains almost no blue .
how a computer find edges once you maths on them .
the most important operation in computer vision is the filter. a small grid of weights that slides over the images. at each position you multiply the pixels under filter by its weights and add everything up.
a famous example is the solber filter:

-1 0 1
-2 0 2
-1 0 1
it subtracts the left side
from the right side . look at what happens on two small patches:
. a flat patch where every pixel is 1000: the left side gives -400, the right side gives +400 and the total is 0 . no edge.
.an edge patch where the left is 0 and the right is 200: and the right is 200: the left give 0, the rigt gives +800, and the total is 800. a bright gives edge.
so the filter light up exactly where brightness change sharply, whaich is what an edge is. flip the filter on its side and it finds horizontal edges instead. other filters blur or sharpen an image. the only difference between them is the numbers inside.
From hand-made filters to CNNs

Someone designed the Sobel filter by hand. A convolutional neural network (CNN) does the same sliding-and-adding, but nobody designs the filters. The network learns them from labelled examples using the same training loop as any other neural network.

Researchers who look inside trained CNNs find a neat pattern: the first layers learn edges and colours, the middle layers learn textures and parts such as eyes or wheels, and the last layers respond to whole objects.

Pixels, then edges, then textures, then objects.

Five core tasks
Classification: one label for the whole image.
Detection: a box around each object.
Segmentation: a label for every single pixel.
OCR: reading text in images.
Tracking: following objects across video frames.

In my notebook I ran three of these on the same street photo using small pretrained YOLO models. The classifier gave one label, “minibus”, at 57% confidence. That is not wrong, but it is not very useful for a busy street. The detector found the bus and four people with boxes. The segmenter outlined exactly which pixels belonged to each object. More detailed output needs more detailed training labels, and those cost more to create.

YOLO stands for You Only Look Once. Older detectors first proposed possible regions and then classified each one, which was accurate but slow. YOLO looks at the whole image in a single pass, which is fast enough for live video and small devices.

Why is computer vision booming now?

Five things came together:

Cameras everywhere. Every phone is a camera and a small computer.
Data. Huge image collections and satellite imagery.
Compute. Cloud GPUs for training, and small models that run on phones.
Pretrained models. You do not start from zero. You can adapt an existing model with a few thousand local photos.
Multimodal AI. Assistants can now read images as well as text. A Vision Transformer even cuts an image into 16 × 16 patches and treats each patch like a word.
Computer vision across Africa

This is the part that excites me most. Some examples from the course material:

Agriculture: Nuru helps farmers diagnose cassava diseases from a leaf photo, offline, on a basic phone.
Health: In Nigeria, software scored chest X-rays to decide who should get a lab test for tuberculosis. In one project covering 5,298 people, 85 TB cases were found.
Conservation: In Gabon, smart camera traps spot elephants and people and alert rangers by satellite within minutes.
Finance: Smile ID matches a selfie to an ID photo and checks that it is a real, live person.
Mapping: Google’s Open Buildings dataset outlines about 1.8 billion buildings, helping with vaccination planning and disaster response.
Accessibility: BLAIVA, built by FabLab Rwanda and still in beta, is a wearable camera that describes objects, text and faces through earphones.

That last one was built in Kigali. It is a reminder that the people reading this can build these systems too.

When models fail

A model only knows what it was trained on. To see this, I made the street photo harder in three ways: very dark, upside down and very low resolution. These are everyday conditions with cheap phones, poor lighting or a dusty lens. The detector’s results changed in each case. In one it even invented an object that was not there.

This matters because most popular models were trained mostly on photos from Europe, North America and Asia. They can struggle with African crops, clothing, road scenes, skin tones and lighting. For a face-verification system, a missed match is an inconvenience. For a medical screening tool, a wrong answer could harm someone. The fix is local data and fine-tuning, and being honest about how a model was test
Try it yourself

You only need a free Google Colab account:

Load a photo with Python and print its shape. Notice that it is height × width × 3.
Print the red, green and blue values of one pixel.
Pick the 10 × 10 patch with the most contrast and look at its numbers.
Run a Sobel filter and watch the edges appear.
Run a pretrained YOLO model on your own photos, and keep one hit and one miss.

Try a photo of your street, a market or a plant leaf. Where does the model succeed, and where does it fail?

Final thoughts

Computer vision looks like magic until you see what is inside: images are numbers, filters find patterns in those numbers, and networks learn which patterns matter. Once you understand that chain, you can start building tools that solve problems close to home, from crop disease to accessibility.

The next time you glance at a photo and instantly know what it shows, remember that a computer needs millions of numbers and a lot of training to do the same.

words embeddings

Inezayimana Henriette — Wed, 30 Sep 2026 14:57:00 +0000

words embeddings are a way of representing a words as numbers(a vectors) so that the computer can understand the relationship and similalities between words,
the vectors usually 50 to 300 numbers long. These numbers are learned from large amounts of text, and they place words with similar meanings close to each other.
why words embedding is matters to help computer to understands human languege becouse computes hear only numbers. ex on map, cities that are near each other are similar in location. In an embedding space, words that are near each other are similar in meaning. "Doctor" ends up close to "nurse" and "hospital", and far from "bride".
the simplest way is bag-of- word is a way to converting text into a numbers so a computer can process it where each words gets its own position in a long list ,and we count how is appears. example "i love my school" on this computer convert it like ,i:1, love:1,my:1 school:1 [1,1,1,1} this what computer receive , related method is tf-idf,which gives more weight to rare, informative words. both work, but they have a big weakness:they dont understand meaning becouse they count words they dont care the meaning of words.
in bag-of-words, "good" and "great" are as different as "good" and "banana". It treats every word as a separate item with no connection to any other.
how word2vec works on high levels
first Word2Vec was created at Google in 2013.it trains a small neural network on task like predicting words and find their neighbours
and also the words embedded its capture meaning of the word not spelling and also a model trained on one tsk can reuse embedding that were learned from high amount of text even if your dataset is small it help you for that
how it work

they look at the neghbours: it read a lot of text and slides asmall window across eachStep 1: Look at the neighbours. Word2Vec reads a lot of text and slides a small window across each sentence. Take the sentence:"I drink hot coffee every morning"For the word coffee with a window of 2, the neighbours are: drink, hot, every, morning. These pairs (coffee → drink, coffee → hot, and so on) become the training examples. 2: Give every word random numbers. At the start, each word gets a vector of random numbers. These numbers mean nothing yet. 3: Practise a prediction game. A small neural network tries to guess words from their neighbours. There are two versions of the game: v1:continous bag of words they try to predict the middle word colles the target words ex "i love cuckies " the target will be the "love" so you have to find love inorder to fill a sentences "i -cockies" they have need love to fill the sentence this it predit it v2.skip-gram they trie to predict the sarrounding words you get a middle input ex :i love this it find cockies depend on "i love cockies" CBOW is faster compare skip-gram works better for rare words 4: Learn from mistakes. Each time the network guesses wrong, it adjusts the word vectors slightly. After millions of sentences, words used in similar contexts end up with similar vectors. Coffee and tea both appear near drink, hot and cup, so their vectors move close together. 5: Keep the vectors, throw away the game. We don't care about the predictions. The learned vectors are the useful result, and they are the word embeddings. on training model i got code from chart gbt from gensim.models import Word2Vec

sentences = [
["i", "drink", "hot", "coffee", "every", "morning"],
["she", "drinks", "hot", "tea", "every", "morning"],
["he", "drives", "a", "car", "to", "work"],
]

model = Word2Vec(
sentences,
vector_size=50, # length of each word vector
window=2, # how many neighbours to look at
min_count=1, # keep all words
sg=1 # 1 = Skip-gram, 0 = CBOW
)

print(model.wv["coffee"]) # the vector for "coffee"
print(model.wv.most_similar("coffee")) # closest words
what i learned and what suprise me
first what it`s suprise me how computer understand our sentences or words in mathematics lic to convert it with encording as numbers and decording as a word or answer that make a sence i also learn the method of embedding called word2vec how it work depend on what they look at the neighbour of the word lik i drink hot coffee and give word a random number and even how it train it

what supprise me is how you can do a math with a words
i run this analogy
king -man+woman=Qween
the model was never taught what a king or a queen is. It only read a lot of text, and it still found the right answer. I didn't expect a computer to pick up relationships like this on its own