Skip to main content

How to differentiate between a Zeelander 44, 55 and 72?

One day, your life might depend on being able to tell the difference between a Zeelander 44, a 55 and the 72.

At first glace, these boat models are quite similar to each other. You could easily think that a Z44 is a Z55 closer to you.

Three different models. Which is which?
To figure this out, you can spend 5 hours manually reviewing and annotating a thousands photos, deleting bad training data, and staring at tiny thumbnails photos of these boats until your eyes start bleeding.

Then, put this data into a ResNet34, and see that after all of that effort you are able to tell each model from the others 60% of the time. In other words, it guesses correctly less than every two out of three attempts.

A ResNet34 can tell these models from each other 60% of the time.

That's crap. A waste of time and money.

Here is how you differentiate these models like a human.

The Z44 has windows in sets of two, the Z55 has windows in sets of three, the Z72 has an uneven window layout, with four large windows in the center, two at the front, and a vent at the back.

When you need to name the model of the boat while flying over it in a helicopter, observe the front deck: the Z44 has a round cover/window, the Z55 has three dark covers, the Z72 has a rectangular cover.


How to tell a Z44 from a Z55 from a Z72 in a glance.

These two rules alone should allow you to tell a Z44 from a Z55 and from a Z72, at least 80% of the time. I hope you don't stumble onto the remaining 20%, or bad things will happen to you, for sure.

I hope you learned something useful today.

Comments

Popular posts from this blog

Basic cell counting and segmentation in Matlab

Counting cells manually is a tedious error prone process for humans. Given a large data set of microscopy images this task can be achieved much faster by means of basic computer vision techniques. In this tutorial we will segment cells from an image following a method similar to the one presented by Yongming Chen in 1999. The method uses basic morphological operations and the watershed algorithm to segment the cells. Nowadays better methods for cell segmentation exist. This method was chosen for its simplicity and ease of implementation.

We start with an image of cell-like structures by Anna-Katerina Hadjantonakis and Virginia E Papaioannou.

A = imread('cells.jpg');
We convert the image to grayscale:
I = rgb2gray(A);

To be able to extract the dimmer cells, it is necessary to perform some local contrast adjustments
I = adapthisteq(I);


Objects on the borders can be caused by noise and other artifacts. We can eliminate objects on the borders of the image like this:
I = imclearborder(…

Project planning in a text file

Whenever you work on a project it is important to be able to plan it ahead of time. This holds true for small and big project, from planning a trip to the spa to building a spaceship. The small project plans can be maintained in you thoughts while bigger ones require tools to help you see the big-picture of the project and manage task at a lower level. There are projects which start with a fully prepared plan and projects which pivot overnight, thus invalidating any original plan. For the latter flexibility is very important, and tools like Trello offer a great solution because they can be adjusted to fit your project.

However, it may happen sometimes that the project starts adjusting to the tool or that you still want to maintain a bigger picture of the main points of the project. You may also need to produce a rough development schedule to serve as a long term road-map.

I have prototyped a tool (and defined a workflow) which allows you to plan such projects.

To better understand how…

A more confusing Internet with .brand TLDs

A few months ago, I wrote a post proposing to remove top level domains from the internet, as many people don't understand why we need TLDs. At the time I didn't know either.
Removing TLDs would simplify browsing an possibly increase security by reducing phishing sites that make use of URL similarity to gain visits to potentially dangerous websites, or sites filled with ads.
The ICANN, the body that manages Internet names, has for years been preparing something different but that produces a slightly similar effect than removing TLDs: allowing brands to register their own TLDs. The proposal may soon be approvedhas been approved.
Brands would therefore be able to register their own TLDs. Google could register .google a potentially powerful TLD for their services, like Google Maps: maps.google instead of maps.google.com.
Brand TLD don't improve short urls like google.com, but seem to be effective for use in subdomains.
Registering a .<brand-name> TLD will cost $185K, wh…