I want to get my hands dirty with some machine learning, and I finally have a problem which seems like a good beginner project. However, despite reading a lot about the subject I am unsure how to get started, and what my basic approach should be.
I have a dataset which should look like this.
a real dataset looks more like this:
I want to identify the points in the red circles (on the first image), and be robust against occasional artifacts like the one in the blue circle.
I sounds like a really easy task. However, the is quite a lot of noise in the raw data. My current implementation is pretty traditional. It blurs the data and compares the first and second derivative to some estimated threshold values. This approach works, but can "only" identify the points with ~99.7% accuracy, but since I do around 100.000 measurements a day I would love to increase this number.
So, this is what I have:
All the datasets I want/need
A pretty good model of how the data should look.
A pretty good training set, using my existing algorithm (the outlines can be fixed manually)
However, I do not have a basic idea how what approach I should use. I feels like none of the material I've read on machine learning fit's this problem.
Can someone help me with the super high level approach to solve this problem?
Related
I would like to draw a transit map which is not based on any real map.
Unlike conventional maps, transit maps are usually not geographically accurate—instead they use straight lines and fixed angles, and often illustrate a fixed distance between stations, compressing those in the outer area of the system and expanding those close to the center.
This map would be massive, not infinite but if a line ran horizontally across it could ideally have 40,075 stations. I want it to look just like any local transit map (I'm basing myself on the Montreal metro map) but much bigger obviously, which means I don't care about what a metro system of this scale should look like or how useless a map this size would be.
I think the hardest part will be to generate where the stations will be, then drawing stylized lines between those stations should be relatively easy using something like Processing.
So, do you have any idea how to generate a giant transit map???
So far, the research
Nathan Hellinga's Processing.py subway map generator resembles what I'm looking for and looks great but the algorithm wouldn't scale well to a very large grid.
Jannis Redmann's generating transit map theory is really interesting but bases itself on real world data. Maybe it could be used with generated data but how do you generate that data then?
My idea, a random walker
Basically, roll some dice and based on a predefined set of rules: go forward, place a station, turn... and repeat countless times until the map is filled. I'm not yet sure what the probabilities would be, it would take some trial and error.
Results of another question I asked on Worldbuilding
Fractal generators loos really promising! But how do I make it look like a transit map? I think it relates to the slime (see below) so I'll look more into it.
Graphviz, an open-source tool that converts DOT script files into graphical images. I think that has the same problem as Jannis Redmann's, I still need an algorithm to generate data.
Slime is a really interesting idea! I would have to do some more research on how to reproduce these patterns but it's an interesting place to start.
In image processing, each of the following methods can be used to get the orientation of a blob region:
Using second order central moments
Using PCA to find the axis
Using distance transform to get the skeleton and axis
Other techniques, like fitting the contour of the region with an ellipse.
When should I consider using a specific method? How do they compare, in terms of accuracy and performance?
I'll give you a vague general answer, and I'm sure others will give you more details. This issue comes up all the time in image processing. There are N ways to solve my problem, which one should I use? The answer is, start with the simplest one that you understand the best. For most people, that's probably 1 or 2 in your example. In most cases, they will be nearly identical and sufficient. If for some reason the techniques don't work on your data, you have now learned for yourself, a case where the techniques fail. Now, you need to start exploring other techniques. This is where the hard work comes in, in being a image processing practitioner. There are no silver bullets, there's a grab bag of techniques that work in specific contexts, which you have to learn and figure out. When you learn this for yourself, you will become god like among your peers.
For this specific example, if your data is roughly ellipsoidal, all these techniques will be similar results. As your data moves away from ellipsoidal, (say spider like) the PCA/Second order moments / contours will start to give poor results. The skeleton approaches become more robust, but mapping a complex skeleton to a single axis / orientation can become a very difficult problem, and may require more apriori knowledge about the blob.
I understand how to do classification problems and starting to understand convolution networks which I think is the answer to some extent. I'm a bit confused on how to setup a network to give me the output position.
Let's say you have the position of the end point of noses for a data set with faces. To find the end point do you just do a 'classification' type problem where your output layer is something like 64x64 = 4096 points but if the nose is at point row 43 and column 20 of your grid you just set the output as all zero's except for at element 43*64 + 20 = 2772 where you set it equal to 1? Then just map it back to your image dimensions.
I can't find much info on how this part of identification works and this is my best guess. I'm working towards a project at the second with this methodology, but it is going to be a lot of work and want to know if I'm at least on the right track. This seems to be a solved problem, but I just can't seem to find how people do this.
Although what you describe could feasibly work, generally neural networks (convolutional and otherwise) are not used to determine the position of a feature in an image. In particular, Convolutional Neural Networks (CNNs) are specifically designed to be translation invariant so that they will detect features regardless of their position in the input image - this is sort of the inverse of what you're looking for.
One common and effective solution for the kind of problem you're describing is a cascade classifier. They have some limitations, but for the kind of application you're describing, it would probably work quite well. In particular, cascade classifiers are designed to provide good performance owing to the staged approach in which most sections of the input image are very quickly dismissed by the first couple stages.
Don't get me wrong, it may be interesting to experiment with using the approach you described; just be aware that it may prove difficult to get it to scale well.
I'm working on a project with many unknowns like moving the app from one platform to another.
My original estimations are way off and there is no way I can really know for sure when this will end.
How can i deal with the inability to estimate such a project. It's not that I'm adding a button to a screen or designing a web site, or creating and app or even fixing bugs. These are not methods with bugs, these are assumptions made in the overall code, which are not correct anymore and are found step by step and each analyzed and mitigated with many more unknowns.
I happened to write a master thesis about software-estimation and there are lessons I've learned:
-1st Count, 2nd compute, 3rd judge - this means: first try to identify items in your work which are countable e.g files, classes, LOCs, UIs, etc. Then calculate using this data the effort (in person/days). Use judgement as the last ressort.
-Document your estimation! Show numbers. This minimizes your risk, thus you will present results not as your opinion, but as more or less objective figures. (In general, the more paper the cleaner the backside)
-Estimation is not a commitment. Commitment is one number, estimation is always a range - so give your estimation as a range ( use cone of uncertainty to select the range properly http://www.construx.com/Page.aspx?hid=1648 )
-Devide: Use WBS, devide your work in small pieces and estimate them separately. The granulity depents on the entire length, but at most a working-package soultn't be bigger than 10% of entire effort.
-Estimate effort first, then schedule, then costs.
-Consider estimation as support for planing, reestimate on each project phase (s. cone of uncertainty).
I would suggest the book http://www.stevemcconnell.com/est.htm which deals all these points, in particular how to deal with bosses, who try to pull a commitment from you.
Regards,
Valentin Heinitz
There's no really right answer for coming up with an accurate estimation, because there's no way to know it.
as for estimating the work itself, think about how each step can be divided into separate sub-steps, and break those down even smaller, until you can get a fair picture of as much of the work as you can, with chunks small and discreet enough to give sound estimates for. If you can, come up with both an expected time and a worst-case time, to get a range of where you could land.
Another way to approach this is to ignore the old system. It sounds like a headache. Make an estimate of scraping the old system and implementing a new one from scratch, or integrating a 3rd party, off the shelf solution. If there's a case to be made for this, it is worth at least investigating it.
Sounds like a post for postsecret not SO. :)
I would tell him that it will be done when its done, and if thats not good enough, he can learn to program and help you. Then again, I think that you might get fired, but hey that sounds like it might be better.
Tell him more or less what you told us. The project is too volatile too give an accurate estimate and the best you can do is give an estimate for a given task. As long as the number of tasks is unknown so will be the estimate. If he is at all worth his salary he would rather hear this than some made up number. This is not uncommon when dealing with a large legacy code base.
It's not that I'm adding a button to a screen or designing a web site,
or creating and app or even fixing bugs.
That is a real problem. You can not estimate what you don't have experience in. The only thing you can do is pad your estimate until you think it is a reasonable amount of time. The more unknowns you think there are the more you pad. The less you know about it the more you pad.
I read the below book and it spoke at length about accuracy vs precision. Basically you can be accurate but have a very large range. For instance you can be certain the task will be between 1 day and 1 year to complete. That is not very precise but it is really accurate.
Software Estimation Demystifying...
Some tips for estimating
I want to write an algorithm that can take parts of a picture and match them to another picture of the same object.
For example, If I gave the computer a picture of a vase and a picture of a scene with the vase in it, I'd expect it to determine where in the image the vase is.
How would I begin to develop an algorithm like this?
The final usage for this algorithm will be an application that for example with a picture of somebody's face could tell if they were in a crowd of people. This algorithm would eventually be applied to video streams.
edit: I'm not expecting an actual solution to this problem as I don't hope to solve it anytime soon. The real question was how do you define something like this to a computer so that you could make an algorithm to do it.
Thanks
A former teacher of mine wrote his doctorate thesis on a similar sort of problem, except his input was a detailed 3D model of something, which he would use to find that object in 2D images. This is a VERY non-trivial problem, there is no single 'answer', certainly nothing that would fit the Stack Overflow format.
My best answer: gather a ton of money and hire a very experienced programmer.
Best of luck to you.
The first problem you describe and the second are both quite different.
A major part of each is solved by the numerous machine vision libraries available. You may need a combination of techniques to achieve any success at either task.
In the first one, you would need something that generically recognizes objects. Probably i'd use a number of algorithms in concert to identify the foreground object in the model image and then do some kind of weighted comparison of the partitioned target image.
In the second case, examining faces, is a much more difficult problem relative to the general recognizer above. Faces all look the same, or nearly so. The things that a general recognizer would notice aren't likely to be good for differentiating faces. You need an algorithm already tuned to facial recognition. Fortunately this is a rapidly maturing field and you can probably do this as well as the first case, but with a different set of functions.
The simple answer is, find a mathematical way to describe faces, that can account for angles and partial missing data, then refine and teach it.
Apparently apple has done something like this, however, it still makes mistakes and has to be taught as it moves forward.
I expect it will be more about the math, than about the programming.
I think you will find this to be quite a challenge. This is an extremely difficult problem and is one of the many areas of computing that fall under the domain of artificial intelligence (AI). Facial recognition would certainly be the most popular variant of this problem and in spite of what you may read in the media, any claimed success are not what they are made out to be. I think the closest solutions involve neural nets and they require very clear and carefully selected images usually.
You could try reading here though. Good luck!