33
0

User

## Coins classifier Neural Network: Head or Tail?

14 min
1.3K

Home of this article: https://robotics.snowcron.com/coins/02_head_or_tail.htm

The global objective of these articles is to build a coin classifier, capable of scanning your pocket change and find rare / valuable coins. This is a second article in a series, so let me remind you what happened earlier (https://habr.com/ru/post/538958/).

During previous step we got a rather large dataset composed of pairs of images, loaded from an online coins site meshok.ru. Those images were uploaded to the Internet by people we do not know, and though they are supposed to contain coin's head in one image and tail in the other, we can not rule out a situation when we have two heads and no tail and vice versa. Also at the moment we have no idea which image contains head and which contains tail: this might be important when we feed data to our final classifier.

So let's write a program to distinguish heads from tails. It is a rather simple task, involving a convolutional neural network that is using transfer learning.

Same way as before, we are going to use Google Colab environment, taking the advantage of a free video card they grant us an access to. We will store data on a Google Drive, so first thing we need is to allow Colab to access the Drive:

Читать далее
0

## Coins Classification using Neural Networks

19 min
2.8K

See more at robotics.snowcron.comThis is the first article in a serie dedicated to coins classification.Having countless "dogs vs cats" or "find a pedestrian on the street" classifiers all over the Internet, coins classification doesn't look like a difficult task. At first. Unfortunately, it is degree of magnitude harder - a formidable challenge indeed. You can easily tell heads of tails? Great. Can you figure out if the number is 1 mm shifted to the left? See, from classifier's view it is still the same head... while it can make a difference between a common coin priced according to the number on it and a rare one, 1000 times more expensive.Of course, we can do what we usually do in image classification: provide 10,000 sample images... No, wait, we can not. Some types of coins are rare indeed - you need to sort through a BASKET (10 liters) of coins to find one. Easy arithmetics suggests that to get 10000 images of DIFFERENT coins you will need 10,000 baskets of coins to start with. Well, and unlimited time.So it is not that easy.Anyway, we are going to begin with getting large number of images and work from there. We will use Russian coins as an example, as Russia had money reform in 1994 and so the number of coins one can expect to find in the pocket is limited. Unlike USA with its 200 years of monetary history. And yes, we are ONLY going to focus on current coins: the ultimate goal of our work is to write a program for smartphone to classify coins you have received in a grocery store as a change.Which makes things even worse, as we can not count on good lighting and quality cameras anymore. But we'll still try.In addition to "only Russian coins, beginning from 1994", we are going to add an extra limitation: no special occasion coins. Those coins look distinctive, so anyone can figure that this coin is special. We focus on REGULAR coins. Which limits their number severely.Don't take me wrong: if we need to apply the same approach to a full list of coins... it will work. But I got 15 GB of images for that limited set, can you imagine how large the complete set will be?!To get images, I am going to scan one of the largest Russian coins site "meshok.ru".This site allows buyers and sellers to find each other; sellers can upload images... just what we need. Unfortunately, a business-oriented seller can easily upload his 1 rouble image to 1, 2, 5, 10 roubles topics, just to increase the exposure.

So we can not count on the topic name, we have to determine what coin is on the photo ourselves.To scan the site, a simple scanner was written, based on the Python's Beautiful Soup library. In just few hours I got over 50,000 photos. Not a lot by Machine Learning standards, but definitely a start.After we got the images, we have to - unfortunately - revisit them by hand, looking for images we do not want in our training set, or for images that should be edited somehow. For example, someone could have uploaded a photo of his cat. We don't need a cat in our dataset.First, we delete all images, that can not be split to head/hail.

Читать далее
0

## Robotic Floor Washer

16 min
1.8K

When we think about robots, the first thing that comes to mind are robotic vacuum cleaners. The reason is simple: they are the most "solid" demonstration of success of "consumer" robotics. So making one sounds like a good idea... at first.

But isn't it a bit counter productive - to build something that popular, something we can buy in a store at a commodity (small) price? Should we build something similar, but NOT a vacuum cleaner? Something like... a floor washer, perhaps? Yes, a robotic floor washer.

In this tutorial I am going to build a fully working prototype of a robotic floor washer. By "fully working" I mean that it is going to wash floor, instead of moving dirt around like most robotic "moppers" do. While by "prototype" I mean it is going to be the first step towards production-ready unit, but not a production-ready unit yet. Let me explain.

First of all, it is not going to be THAT solid. You can grab a robotic vacuum cleaner that you got from the store by any part, including wheels and bumper and lift it. It will not fall apart. Ours probably will. The reason is, to make a device "mechanically solid" is a separate task, and if we focus on it, then "robotic" tasks will become more difficult to achieve. So we are going to do what engineers usually do: first they build C3PO without the outside body, wires everywhere and so on. And only then they put a gold-covered outfit on it.

Читать далее
+12

## Brainless Platform for a Robotic Vacuum Cleaner

11 min
1.2K
More on robotics.snowcron.com

First of all, this is just an exercise, useful as is, but the result is going to be far from an industrial level robots. Why doing it then?

For the same reasons we do all exercise: to get an experience. After all, when we write a character recognition «MNIST classifier» neural network, we know that the problem is solved long time ago. But we need to become familiar with tools and approaches. Same here.

Now, why is it called «brainless»?
+8

## Testing Water Melon using Neural Networks: Full Dev. Cycle from prototyping to the App. at Google Play

7 min
681

### The beginning

It all started when I found an app. on Apple market, that supposedly was able to determine the ripeness of a water mellon. A program was… strange. Just think about it: instead of knocking using your knuckles, you were supposed to hit the water mellon with your iPhone! Nevertheless, I have decided to repeate that functionality on an Andtoid platform.
Read more →
0

## Определяем породу собаки: полный цикл разработки, от нейросети на Питоне до приложения на Google Play

27 min
22K
Прогресс в области нейросетей вообще и распознавания образов в частности, привел к тому, что может показаться, будто создание нейросетевого приложения для работы с изображениями — это рутинная задача. В некотором смысле, так и есть — если вам пришла в голову идея, связанныя с распознаватием образов, не сомневайтесь, что кто-то уже что-то подобное написал. Все, что от вас требуется, это найти в Гугле соответствующий кусок кода и «скомпилировать» его у автора.

Однако, все еще есть многочисленные детали, делающие задачу не столько неразрешимой, сколько… нудной, я бы сказал. Отнимающей слишком много времени, особенно если вы — новичок, которому нужно руководство, step-by-step, проект, выполненный прямо на ваших глазах, и выполненный от начала и до конца. Без обычных в таких случаях «пропустим эту очевидную часть» отговорок.

В этой статье мы рассмотрим задачу создания определителя пород собак (Dog Breed Identifier): создадим и обучим нейросеть, а затем портируем ее на Java для Android и опубликуем на Google Play.

Если вы хотите посмотреть на готовый результат, вот он: NeuroDog App на Google Play.

Веб сайт с моей робототехникой (в процессе): robotics.snowcron.com.
Веб сайт с самой программой, включая руководство: NeuroDog User Guide.

А вот скриншот программы:

Читать дальше →
+16

## Dog Breed Identifier: Full Cycle Development from Keras Program to Android App. on Play Market

25 min
16K
With the recent progress in Neural Networks in general and image Recognition particularly, it might seem that creating an NN-based application for image recognition is a simple routine operation. Well, to some extent it is true: if you can imagine an application of image recognition, then most likely someone have already did something similar. All you need to do is to Google it up and to repeat.

However, there are still countless little details that… they are not insolvable, no. They simply take too much of your time, especially if you are a beginner. What would be of help is a step-by-step project, done right in front of you, start to end. A project that does not contain «this part is obvious so let's skip it» statements. Well, almost :)

In this tutorial we are going to walk through a Dog Breed Identifier: we will create and teach a Neural Network, then we will port it to Java for Android and publish on Google Play.

For those of you who want to see a end result, here is the link to NeuroDog App on Google Play.

Web site with my robotics: robotics.snowcron.com.
Web site with: NeuroDog User Guide.

Here is a screenshot of the program:

Read more →
+11

## Определяем спелость арбуза с помощью Keras: полный цикл, от идеи до программы на Google Play

8 min
38K

### С чего все началось

Все началось с Эппл Маркета — я обнаружил, что у них есть программа, позволяющая определить спелость арбуза. Программа… странная. Чего стоит, хотя бы, предложение постучать по арбузу не костяшками пальцев, а… телефоном! Тем не менее, мне захотелось повторить это достижение на более привычной платформе Андроид.
Читать дальше →
+62

## Information

Rating
Does not participate
Registered
Activity