Automating bank cheque analysis by using Microsoft Cognitive Services

Just looking for code?

Microsoft Cognitive Services is a rich set of AI services, such as Computer Vision, Speech Recognition, Decision making and NLP. The great thing about these tools is that you don’t really have to be an AI expert to make use of these tools, as these models come pre-trained and production ready. You’ll just feed it your information and let the framework work for you.

We’ll be looking at one area of Microsoft’s Cognitive Services – Computer vision. More specifically, we’ll be looking at the handwriting API – you’ll provide the handwriting and the system will provide you with the actual text. We’ve already worked with the Computer Vision API from Microsoft Cognitive services – we used this API to tag our photo album.

Let’s look at today’s scenario – we’re a fictitious bank which processes bank cheques. These cheques come hand-written from our clients, which contain instructions on how to transfer money from one account to the holder’s account.

A cheque typically has the following information:

  • Issue Date
  • Payee
  • Amount (in digits)
  • Amount (in words, for cross reference)
  • Payer’s account number
  • (Other information, which was omitted for this proof of concept

This is how our fictitious cheque looks like.

Fictitious Cheque
Ficticious cheque – credit:

This is how our fictitious cheque looks when we’re looking at the regions we’re interested in, represented in bounding boxes.

Cheque Template with bounding boxes representing areas of interest

Let’s consider these three handwritten cheques.

The attached application does the following analysis:

  • Import these cheques as images.
  • Send the images over to Microsoft Cognitive Services
  • Extract all the handwriting / text found in the image
  • Consider only those text which we’re interested in (as represented with bounding boxes previously)
  • Forward this extracted information to whatever system needed. In our case, we’re just printing them to screen.

The below is the resultant information derived from the sample cheques.

Results of Cheque Analysis

Most of the heavy lifting is done by the Microsoft Cognitive Services, making these AI tools available to the masses. Of course, with a bit more business logic, the information that can be extracted from these tools can be greatly improved, making them production ready.

As with the previous example, this example uses the TPL Dataflow library, which is an excellent tool for Actor-Based multithreaded applications.

If you want to try this yourself, you’ll need:

Until the next one!

Watch videos with your friends even if you’re under quarantine or lockdown at home!

Let’s face it – staying at home under quarantine and lockdown due to COVID-19 is not fun at all. You might want to watch some videos in sync with your loved ones. You can take the silly and manual way and say – 3,2,1, PLAY! That may work, but if someone needs to pause, you’ll need to go through that silly 3,2,1 again! What if there are tools to help you out?

Turns out, there is Syncplay Syncplay is a free tool that allows you to synchronize what video you’re watching, even if you are miles away. It works with modern playback tools; I’ve tested this tool with VLC and it was great.

Let’s set it up! -Firstly, go to and download your version. In my case, we’ve tried it on both Windows and Mac and it was great. Once you’ve downloaded it, fire the app up. You’ll need a server so that everyone can connect and play. Syncplay provides free servers to connect to, here’s a list:

  • pl:8995
  • pl:8996
  • pl:8997
  • pl:8998
  • pl:8999

Fill out the following fields:

  • Server Address -> Pick one from the list above
  • Username -> Just your name so your friends can know who is who
  • Default Room -> This should be unique so that only your friends can join. Just pick a fancy name which is unique to you.
  • Path to media player -> Here, we’re telling Syncplay which application to use to play videos. I’m using VLC – find the path of the application and save it.
  • I also recommend “Enable shared playlists” but this is completely optional.
Syncplay – configuration and connection

Once you’re done – hit “Store configuration and run Syncplay”. If all goes well. Syncplay will show the below screen, and your media player should start. In my case, we’re two people in the room.

Two people int the same room

To start playing a video, go to Syncplay -> File -> Open Media File -> and navigate and load your file. Make sure that everyone loads the same file with the same name. You’ll get a notification if the file is not the same, such as the below.

File name, size and duration is not the same; you’ve loaded a different file.

If the same file name with the same length is opened, you’ll see the below.

Files are the same – just waiting for people to be ready

Once you’re ready, from Syncplay, press I’m ready to watch!

Checkbox to show you’re ready

When everybody is ready, you can proceed and play! Any one person can play and pause video for all the other users in the room. Enjoy synchronized playing!

You can also comment during playback and it will appear in both VLC and Syncplay – it’s like you’re almost next to each other!

Chat appearing both in VLC and Syncplay

There are many other features you can explore in Syncplay:

  • Loading and Sharing a Media Directory
  • Loading a network stream
  • Creating Playlists

Enjoy and stay safe during these turbulent times!

How to avoid getting your credit card details stolen online (some practical ways)

Let’s immediately jump through the points, then some back-story.

  • Only submit credit card details to make a purchase on shops which are super famous, such as Amazon, eBay, official product merchants
  • Use 3rd party paying capabilities where possible, such as PayPal. With this method, the merchant never has access to your credit card details, only to the authorized funds.
  • Ideally use a debit card rather than a credit card online. In case your card gets stolen, the thief can only use the available funds, without ending up in debt.
  • Also, only transfer the money your’re going to use when making the purchase. The card should empty in general.
  • Use applications such as Revolut which have capabilities to enable / disable online transactions on demand.
  • Revolut also allows for disposable credit cards – the number changes every time you make a transaction. This means that even if your credit card is stolen, the card is now dead and worthless. You’ll need to pay a monthly fee, though.
  • Avoid saving credit card details on your browser. Although usually CVV is usually not saved, writing it down again won’t take long. This avoids the possibility of a virus sniffing down your credit card details if they are saved.

Why am I writing these? I’ve just stumbled on a Cyber Security Episode done by MITA / Maltese Police (great initiative, by the way). Although the content made sense, I felt that some practical points were missing (original video here). The premise of the video is to only buy from sites using HTTPS as it’s secure and you’ll know the seller. Some points on the premise:

  • Running HTTPS ONLY GUARANTEES that the transmission between you and the merchant is secure. It does NOT mean you truly know who’s responsible as a merchant. There are ways to “try” and fix this (through EV certificates) but EVs are probably dead as well.
  • What happens after your credit cards are securely transported is unknown. The following may occur:
    1. Merchant might be an outright scammer running a merchant site with an SSL certificate (which nowadays, can be obtained for free,
    2. Merchant might store your credit card details insecurely; he may end up getting hacked and credit card details will get stolen.


The BOV PAY MVP Problem

BOV has a problem – Revolut is seriously eating through its card payments business (and profits). Revolut in Malta is very famous and for a very good reason. No, it’s not because it provides free, VISA / Chip / Contactless cards; BOV does that as well. It’s because Revolut’s mobile app is just awesome. BOV seems to (finally) notice this and they have Revolut at their sights. How can BOV one-up Revolut? Tap-to-pay purchases!

BOV have been working on their now BOV PAY app – this will enable a mobile user to make purchases by tapping their phones. Mobile vendors have already developed this though; the likes of Google Pay, Apple Pay, Samsung Pay already exist. But here’s the problem – in Malta, only Apple Pay is officially supported, so non-android users are hung out to dry (unless you follow my unofficial guide here)

Smartly enough, BOV prioritized launching an app for Android firstly. The reason is simple – since there is no official tap-to-pay application in Malta, they’re tapping (pun intended) into an untapped market. So, android users, rejoice!..or should we?

You see, the application is in a very early phase; I’d say it’s just an MVP. The application is sluggish, the user experience is quite rubbish and the app is riddled with bugs. Tap-to-pay works, but the experience is frankly rubbish. Once I tapped my phone to pay, the app took more than 5 seconds to simply show the BOV screen and the card that I’ve paid with. But alas, it works.

Let’s discuss the next most infuriating thing. Honestly, once they get this (properly) working, I’ll seriously consider switching back to use BOV cards instead of Revolut cards daily. There is NO WAY to INSTANTLY see the transactions that I’ve affected!

Well, firstly, the new BOV PAY app does not show any transaction whatsoever. I’m guessing that this is a work in progress, and I should be using the older app to check the transaction. OK, fine. But, still, the old app does not enable me to INSTANTLY see my transaction. I’m able to see that there is a difference between book balance and available balance, but the transaction does not show up. This is PAINFUL – show me that I’ve made a payment, even though the payment may be reverted if unclaimed by merchant. It’s my right to see where my money went! UGH!

Anyway, let’s dive into some other criticism about the app; in no order:

  • It takes at least 5 seconds to get the BOV logo and the loading spinner
  • It takes another 5 seconds (or more) from successfully recognising my fingerprint into the main menu
  • I had to add all my personal details and sign up. Why did I had to provide my name, address and other information? Can’t this be derived from ownership of the card? Or maybe through the authentication of the classic BOV app?
  • When I press back, I prompted to log out; why? Isn’t it obvious that I’m the only person who’s using this app? Logout should be buried somewhere; I don’t need to see it every time.
  • The card controls feature is very weak – it only has ONE option. I’m guessing that this is WIP – no harm, no foul.
  • Again, I’m not sure why through the Review Card Details I’m able to edit information such as the expiry date, CVV, address. This needs to be done the other way round; I’d prove who I am and cards are added automagically for me and the relevant information is obtained from your records. This is a VERY WEIRD feature.
  • From a UI Experience. why does load the picture of the card each time? I’m always looking at a spinner before the card image is displayed.
  • Expiring sessions? Really? What’s protecting me from? As soon as the app is dismissed, THEN and only THEN the session should be terminated.
  • The locator feature; I’m assuming that this shows a list of ATMS? Poor naming?
  • The setting section looks very basic; but I guess it gets the job done. OK
  • The about us is very cheeky feature; it’s just a WebView of the BOV Site. Lazy!
  • Tutorials
    • The Pay-In-Site is a good and easy tutorial; the button to go to the cards section doesn’t work.
    • Using the App is again, a WebView of the BOV site.
    • Contact us – again, WebView.
  • Terms and Privacy – All WebViews (oh, did I mention that the site is NOT mobile friendly?)

I also must mention the fact that I’ve forwarded much of the feedback to BOV and they instantly got back to me over the phone to listen to my feedback. I must give credit where it’s due. I’m hoping that they the feedback that the clients are giving to them so that they can create an awesome app.

They’ve started on the right track, but they got so much progress to do. I’m just hoping that eventually they’ll introduce the Google Pay integration. But this will have to do for now. I’ll be sticking to my Google Pay and Revolut combination though; it works great.

EDIT – Here’s a demo of the BOV PAY in real life – note how long it takes for the BOV APP to trigger (and show up) after payment takes place.

Store your MySQL Docker database info in a Docker Volume

If you spin a MySQL Docker container, you’ll notice that once the container is stopped, all the information is lost! In order not to lose any information from your MySQL Docker database, a volume will need to be attached to the container. Let’s do that!

Let’s create a new volume – this will be used to store all your database informaton

docker volume create mysql

Once the volume is created, it can be attached to a newly spun MySQL container

docker run --name mysql -e MYSQL_ROOT_PASSWORD=albert --mount source=mysql,target=/var/lib/mysql -d mysql

Any datatabses created on this MySQL instance is now preserved! Let’s test it out. Let’s connect to the container and create a new database:

docker exec -it mysql mysql -uroot -p

Supply your password (in this example, the password would be the value that we’ve supplied for the MYSQL_ROOT_PASSWORD – albert

Once you’re connected, let’s see the current databases. The default installation will have 4 default databases

| Database           |
| information_schema |
| mysql              |
| performance_schema |
| sys                |

Let’s go ahead and create a new database!


If you get all the databases now, you should get the following

| Database |
| information_schema |
| mysql |
| performance_schema |
| sys |
| test |

De-attach from MySQL – type exit. Let’s now remove the current container and re-create a new container; re-attach the previously created volume.

docker container rm mysql -f
docker run --name mysql -e MYSQL_ROOT_PASSWORD=albert --mount source=mysql,target=/var/lib/mysql -d mysql

Let’s re-attach and get the list of databases (don’t forget to supply the password):

docker exec -it mysql mysql -uroot -p

The output should now read as follows – the database test still exists, even after deleting and re-creating the container

| Database |
| information_schema |
| mysql |
| performance_schema |
| sys |
| test |

How to get your first Docker container up and running on Ubuntu in 2 minutes or less

Looking to install Docker and to play around a bit? If you follow the main guide on the Official Docker site, you’ll be surprised how many steps need to be taken to get Docker Installed.

Thankfully, if you keep scrolling, you’ll notice that there is a convenience script! That’s great! Let’s get Docker installed!

Make sure you got cURL installed:

sudo apt-get install curl

Then let’s use the convenience script to install Docker!

sudo curl -sSL | sh

Once the installation is complete, let’s get your first container running!

sudo docker run hello-world

This should show something along the lines of:

Hello from Docker!
This message shows that your installation appears to be working correctly

Then it means that your installation is now complete!

You can already use Google Pay in Malta!

UPDATE: Some users have reported that Revolut was not allowing them to add cards to their Google Pay accounts in Malta. So, your mileage may vary. Users whose cards are already added will still work fine. This was working as intended at the moment of writing.

Unless you have been living under a rock, you’ll know that Apple Pay is now supported in Malta, through Revolut. But little did Android users know that Google Pay has been supported by Revolut for more than a year now!

According to an article on Android Police, around mid-July 2018, Revolut rolled out support for Google Pay. This means that using any NFC enabled Android device, you can start paying locally with your phone!

What is Revolut?

Revolut is an online bank based in the UK – you can get a free debit card to be used locally and comes with a great Android app. It’s the only card that works with Google Pay in Malta – BOV, HSBC, BNF and any other local bank do not support Google Pay, so Revolut is your only choice in this case. You can sign up with Revolut for free here (note that with each successful sign-up, I get a small kick-back, as described here).

How to I get Google Pay?

Easy – using your Android Phone, just search for Google Pay on Google Play:

Can’t find the app on the store, or it says something like “This app is incompatible with all of your devices”? Just download the APK from some APK Mirror, like Of course, side-loading applications can be risky so proceed with caution.


How do I add my card?

Adding the card is very easy – just follow the setup on screen. If you ever used your Revolut card with Google, you’ll find it right there. If not, just add your details. Have your Card No, Expiry Date and CVV at hand though. You’ll need to confirm a code by SMS as well. No screenshots for this section as Google Pay doesn’t allow you to take screenshots when adding cards. You can find a video guide here –

Anything I should know?

Yes – keep in mind that these transactions are considered Online transactions. Therefore, you’ll need to enable Online Transactions in the Card Security section. Don’t worry, if it’s turned off by default, Revolut will warn you about this.


How will my transactions appear?

This is the nice part – they will show up like all other transactions. The payment will not be towards Google; it will point to the retailer you’ve just purchased from.


How can I use Google Pay to pay?

Very easy! Firstly, make sure that NFC is turned on. Then, just unlock your phone and tap the payment machine when the retailer asks you to do so. Your phone will vibrate when you get tour phone close enough and Google Pay will activate itself. You may be asked to verify your identity, such as by supplying a PIN, Fingerprint or lock pattern. By, the way, lock patterns are bad and you should avoid them.

Take that, Apple Pay!

C# Micro Optimizations Part 2 – In Parameter Modifier

In this series of posts, we’re investigating micro-optimizations in C#. As previously mentioned, these may not be applicable to all; but it’s still fun looking at these concepts.

Let’s visit back the last post – Ref arguments. Ref arguments gave us the power of passing structs by value in an extremely efficient manner.

Mutability of a ref struct

Passing structs by ref brings a major disadvantage – the callee might mutate the value of the struct without the caller ever knowing. What if we need to pass structs in an efficient manner, whilst having peace of mind that the callee doesn’t mutate the struct?

Meet the in parameter modifier- C# 7.2

What does the in parameter modifier do? It allows us to pass the argument by reference and giving us the guarantee that the arguments cannot be modified by the callee. Excellent! Let’s run a quick test and make sure our performance is still comparable when passing by ref. Let’s have a struct with 2 properties – let’s have some work done using two different methods – passing by ref and passing by in.

All code can be viewed here –

public class SixteenBitStructBenchmark
    public void BenchmarkIncrementByRef(int limit)
        SixteenBitStruct sixteenBitStruct = new SixteenBitStruct();
        int counter = 0;
            IncrementByRef(ref sixteenBitStruct);
        while (limit != counter);

    public void BenchmarkIncrementByIn(int limit)
        SixteenBitStruct sixteenBitStruct = new SixteenBitStruct();
        int counter = 0;
        while (limit != counter);

    private void IncrementByRef(ref SixteenBitStruct sixteenBitStruct)
        double sum = sixteenBitStruct.D1 + sixteenBitStruct.D2;

    private void IncrementIn(in SixteenBitStruct sixteenBitStruct)
        double sum2 = sixteenBitStruct.D1 + sixteenBitStruct.D2;

public struct SixteenBitStruct
    public double D1 { get; }
    public double D2 { get; }

Let’s see how they perform.

Method limit Mean Error StdDev
BenchmarkIncrementByRef 100000000 23.83 ms 0.0272 ms 0.0241 ms
BenchmarkIncrementByIn 100000000 238.21 ms 0.3108 ms 0.2755 ms

Performance loss?

Wait a second – why is IncrementByIn 10x slower than IncrementByRef when we’re accessing 2 properties in the same struct? Let’s have a look at the generated IL.


 IL_0000: ldarg.1
IL_0001: call instance float64 InOperator.SixteenBitStruct::get_D1()
# Loads argument 1 (SixteenBitStruct) and call the getter


 IL_0000: ldarg.1
# Prepare a new local variable on the evaluation stack
IL_0001: ldobj InOperator.SixteenBitStruct
# Copies the value of SixteenBitStruct into the loaed argument variable
IL_0006: stloc.0
IL_0007: ldloca.s V_0
IL_0009: call instance float64 InOperator.SixteenBitStruct::get_D1()
# Pops the newly created argument into location 0, loads local variable 0 (new copy of SixteenBitStruct) and call the getter

Interesting! When we’ve called the method by ref, the resultant IL just loads the argument and calls the getter. When we’ve called the method by in, the resultant IL creates a copy of the struct before the getter is called. It seems that each time we’re referencing the property, C# is generating a copy of the object for us? We’re facing a by-design feature – a defensive copy.

Why do we encounter a defensive copy?

When calling the getter of our properties, the compiler doesn’t know if the getter mutates the object. Although this is a getter, it’s only by convention that changes aren’t made; there is no language construct that prevents us from changing values in our getter. The compiler must honor the in keyword and generate a defensive copy, just in case the getter modifies the struct.

In the end of the day, a getter is just syntactic sugar for a method. Of course, defensive copies will be generated if methods are called on the struct since the compiles can’t provide any guarantee that the method call won’t mutate the struct.

How do we get around this?

We’ll need instruct the compiler that our struct is immutable, so the compiler doesn’t need to worry about creating defensive copies since values cannot change. C# provides this exact functionality in fact! We can slap the “readonly” keyword (and drop any setters) so that we can guarantee that our struct is now immutable.

Here’s how it looks now

 public readonly struct SixteenBitStruct
    public double D1 { get; }
    public double D2 { get; }

Revisiting our performance numbers

Let’s re-run our benchmarks and assess the performance.

Method limit Mean Error StdDev
BenchmarkIncrementByRef 100000000 23.93 ms 0.1226 ms 0.1147 ms
BenchmarkIncrementByIn 100000000 24.06 ms 0.2183 ms 0.2042 ms

Far better! Performance is now equal (within margin of error). Some closing thoughts about this:

  • Using the in operator is an excellent feature – it allows the callers to safely assume that the values they are going to pass will not have their values changed.
  • Using the readonly modifier with a struct is another excellent feature – it allows the the developer to safely say that its value is immutable and no changes are allowed.
  • The performance uplift is should be considered as a bonus – the design and infrastructure wins using the in / readonly keywords in these context carry far more value.
  • Don’t ever use the in keyword in conjunction with non-readonly structs. Chances are that the performance gained from passing by ref will be lost by accessing the struct’s properties and methods.

Until the next one!

C# Micro Optimizations Part 1 – Ref Arguments

In this series of posts, we’ll be investigating key areas for micro-optimizations. As the title implies, these are micro-optimizations and may not be applicable for you unless you are writing some high-performance library of have a piece of code running in a tight loop. Nonetheless, it’s still fun to investigate and find these micro-optimizations. Onwards!

Let’s start with a simple one – the ref keyword in method arguments. For this argument, we’re only concerned with value type method arguments – structs.

Since structs are value types, by default, the entire struct is copied over to the callee, irrelevant of the size of the struct. If the struct is big, this is typically a bottleneck since a copy must be created and passed for each call. C# provides a method of overriding this behavior by using the ref keyword. If an argument is marked as ref, a pointer to the struct will be passed rather than an actual copy!

This brings two major advantages:

  • If the struct is bigger than 4 bytes (on a 32 bit machine) or 8 bytes (on a 64 bit machine), passing a struct by ref means that less data copying is taking place.
  • We avoid copying back the data – we do not need to return the data since a reference is passed rather than a copy of the struct.

Let’s see an example – lets consider a struct containing two doubles – a 16 byte struct. Let’s say we have two methods that increments one of the values for us (just to give the loop something to do and not get it optimised away).

One of them accepts a (copy of a) struct, increments its internal values and returns the copy back. This is passed by value, which is the default behavior for a struct.

The other method accepts a struct by ref and increments its internal values. There is no need to return the data back therefore no extra copies were needed. This is not the default behavior, so we’ll need to accompany it with the ref keyword.

The below is the source code in question – find the whole solution here:

public class SixteenBytesStructBenchmark
    public void BenchmarkIncrementByRef(int limit)
        SixteenBytesStruct value = new SixteenBytesStruct();
        int counter = 0;
            IncrementByRef(ref value);
        while (limit != counter);
    public void BenchmarkIncrementByVal(int limit)
        SixteenBytesStruct value = new SixteenBytesStruct();
        int counter = 0;
            value = IncrementByVal(value);
        while (limit != counter);
    private void IncrementByRef(ref SixteenBytesStruct toIncrement)
    private SixteenBytesStruct IncrementByVal(SixteenBytesStruct toIncrement)
        return toIncrement;
public struct SixteenBytesStruct
    public long d0, d1;

The below is the time taken for 1000000 runs – this was executed using .NET core 2.2.1 – benchmarks done using BenchmarkDotNet

Method limit Mean Error StdDev
BenchmarkIncrementByRef 1000000 1.663 ms 0.0139 ms 0.0130 ms
BenchmarkIncrementByVal 1000000 2.872 ms 0.0155 ms 0.0145 ms

We can see that running this in a tight loop, doing the work by ref, in this case, is 72% faster! To what can we attribute this performance change? Let’s have a look at what’s happening behind the scenes.

Doing the work by value

Calling IncrementByVal

IL_000a: ldarg.0 # Load the “this” parameter on evaluation stack (implicit)
IL_000b: ldloc.0 # Load SixteenBytesStruct value on the stack (16 bytes worth of data) from location 0
IL_000c: call instance valuetype Ref.SixteenBytesStruct Ref.SixteenBytesStructBenchmark::IncrementByVal(valuetype Ref.SixteenBytesStruct) # Call IncrementByVal with the loaded arguments
IL_0011: stloc.0 #Captures the returned value and stores it in location 0

IncrementByVal Implementation

IL_0000: ldarga.s toIncrement # Load the argument’s address so processing can begin
..method work – removed for brevity
IL_001a: ldarg.1 # Load the value of the field back so it can be returned
IL_001b: ret

What’s happening here?

  • Push the value of SixteenBytesStruct ready to be captured by the upcoming method call
  • Call IncrementByVal
  • IncrementByVal loads the address of the received value from the caller and does the required work
  • Push the value of the SixteenByteStruct after the work has been done ready to be captured by the caller
  • IncrementByVal Returns
  • Pop the value from replace the value of SixteenBytesStruct with the new one

Doing the work by ref

Calling IncrementByRef

IL_000a: ldarg.0 # Load the “this” parameter on evaluation stack (implicit)
IL_000b: ldloca.s V_0 # Load SixteenBytesStruct’s address on the stack (8 bytes worth of data)
IL_000d: call instance void Ref.SixteenBytesStructBenchmark::IncrementByRef(valuetype Ref.SixteenBytesStructamp;) # Call IncrementByVal with the loaded arguments

IncrementByVal Implementation

IL_0000: ldarg.1 # Load the argument so processing can begin. We’re not calling ldarga.s since this already the struct’s address rather than the actual value
..method work – removed for brevity
IL_0018: ret # Return

What’s happening here?

  • Push the address of SixteenBytesStruct ready to be captured by the upcoming method call
  • Call IncrementByVal
  • IncrementByVal gets value received from the caller (the value is an address) and does the required work
  • IncrementByVal Returns

What does this mean?

One can obviously note that doing the work by ref has significantly less work to do:

  • The callee is pushing 8 bytes instead of 16 bytes
  • The callee loads 8 bytes onto the evaluation stack instead of 16 bytes
  • The callee doesn’t need to push the new value onto the evaluation stack
  • The callee doesn’t need to pop the stack and stored the updated value

Therefore, doing the work by ref is pushing less data when a method call takes place (maximum of 8 bytes, irrespective of the struct size) and is avoiding two data copy instructions, since it does not need to push and pop the new value since there are no return values.

If you increase the size of the struct, the performance gains would be even bigger, as shown in the below graph.


We can observe some useful information from this graph

  • When it comes to doing operations by ref, performance is basically equivalent all cross the board, irrelevant to the size of the struct.
  • 16 byte, 8 byte and 4 byte structs carry identical performance – they are just separated by the margin of error.
  • 16 byte, 8 byte and 4 byte structs are faster than 2 byte and 1 byte structs. In fact, 1 byte struct ends up clearly slower than a 2 byte struct! It’s very interesting to explore why 1 and 2 byte structs exhibit performance degradation.
  • The rest of the result show a consistent upward trend – which reflect the amount of data copying take place.

What’s very interesting is that a 4 byte integer operates faster by value when compared to 1 byte and 2 byte integers!

Tag your photo album using Microsoft Cognitive Services

Just interested in the source code? –

Computer vision is one of the key areas that has seen huge growth in both capability and popularity. Though it seems that it’s still out of reach to many; I honestly felt lost when I was trying to play around in this field. It feels like we’re trying to solve problems which have already been solved by other companies. It seems Microsoft shares this vision though, as they’ve introduced Machine Learning features in the form of SaaS.

I’ve stumbled upon Microsoft Cognitive Services through a presentation and I was genuinely amazed. What’s amazing isn’t the results that this service yields – I’ve expected nothing less than excellent results from such tools. What amazed me is how EASY to get involved – there is no fiddling with following pages and pages of guides just to download, install and play around with some software.

Microsoft Cognitive Services enables you to do a huge array of Machine-Learning powered applications, ranging from vision, decision making, natural language processing and other areas. Let’s play around vision – can we use Microsoft’s Cognitive Vision Services and help us organise our photo library?

The idea is that I have many photos, with subjects ranging from food, vacation, family, friends and whatnot. What if my photos contain the proper EXIF tags such as subject and tags? This will allow me to classify my photos by subject and allow me to search through them. What if I can find my photos instantly instead of sifting manually through thousands of photos? I’ll presume that it’s not just me though, everyone has a smartphone nowadays, so this is everyone’s pain.

Great – now we have an objective! Let’s make the tools work for us now. The process will be simple – upload a photo to Microsoft’s Cognitive Vision Services, get the tags and a nice description and slap it to the actual file. Oh, when I say EXIF tags, these can be viewed in File Explorer like below. (Windows 10 Dark Theme in File Explorer here)


Ready to tag your photo library? Let’s go!

Get a Microsoft Cognitive Services Account

Since this is an online service, you’ll need to have an active account with Microsoft Azure. Get your free account from here. Don’t worry, the free service is more than enough to get you playing around. I’ve used the free tier to develop, test and write this blog and I still have plenty of free capacity left.

Create a new Cognitive Services Resource and get the API key

Now that you have an active Azure account, navigate to the Azure Portal and create a new Cognitive Services Resource. Follow the wizard and get the service created – choose whatever region works best for you. I’ve chosen West Europe and the free tier in my case. Once it’s created, we’ll need two things – the URL to our endpoint and our API Key. From the quick start page, get API endpoint and the API Keys.

Get your photos tagged!

Okay, we got all the resources needed, it’s time to get some work done! I’ve created an application to get a photo, upload it our new Cognitive Services resource, get tags and description and apply it to our photo.
Follow these steps to get your photo tagging game going!

  1. Download / Clone my application from GitHub
  2. Open the application and navigate to PhotoAnalyser.cs. Change the subscriptionKey and uriBase to the ones you got previously. The keys in the solution are placeholder keys only.
  3. Run the application – have your photo directory ready as this is asked for at runtime.
  4. Let it do its magic!

In the below example, photo analysis tells us that it’s a pizza on a plate and it also gave us some appropriate tags. Try downloading and viewing the pizza photo -tags and title are preserved as EXIF data.

Keep in mind that the code in the provided solution is not production ready – it’s merely meant as a playground.

Explore the solution

What’s the fun of having a piece of software working without knowing how it works underlying? Here are some points about the application, in no particular order:

  1. It’s making one of the excellent TPL Dataflow framework from Microsoft – this enables the application to scale with ease and to work around the pesky throttling that the free tier carries with it.
  2. It is resizing the images since they don’t need to be large, plus this speed the process up.
  3. It’s using the ImageSharp to resize and add Exif tags to the images.
  4. Given that this application is manipulating images, it is memory intensive. I’ve seen this image hit close to 4GB in memory usage.
  5. It’s split into a library and a consumer just in case.

Continue exploring the Microsoft Cognitive Services stack

Computer vision is just one of the areas in the Microsoft Cognitive Services stack; there are other excellent services to enrich your applications. They also have excellent documentation on this; I’ve followed this to build my application.

That’s it for today! This was an extremely fun project to learn and experiment with new technologies! Until the next one.