[Big Update Available] A.Eye 0.2
So, this is almost after an year I am back!
Lots of things have happened since then, but not much related to A.Eye!
In the last few couple of weeks, I resumed working on A.Eye. Here are the developments.
1) There is a service by name OpenALPR, who has API for number plate detection. It's decently good. Our approach will be to start using this till we have in-house solution!
2) Srinivas Anna won't be fully part of this company. He is relocating to Microsoft US. That is hugely unfortunate for the company. It's kind of removing strong foundation of the company. Now, I will be required to learn things on do it on our own.
3) I have been doing some domain research in last few weeks and realised few things.
1) This is a pretty old field. As old as 1970 where it was used as a surveillance by the police.
2) It required huge setup to run ANPR.
3) There are multiple companies working in this field across different countries and also at home!
4) I haven't come across any big company who is solely working on this project and it's applications. I am sure there will be other startups in this domain, but none of them are yet fully into markets.
Running it alone won't be easy. There are too many things involved. Android, Hardware, AI, Marketing, Cameras and everything else involved in building a startup, including getting funded!
The goal here is to build a platform to get the highest accuracy in ANPR for vehicles in India. The goal is to read any number plate which is readable, with a normal low cost camera connected to internet.
Like any other startup founder, I have no idea why no one else has already done it. I mean, get the highest accuracy on the number plates. And build applications around it.
So, knowingly or unknowingly we are getting into this field called computer vision. I have no idea how far I am from mastering this domain like an expert. Or how much time it takes to do that.
The journey will be starting soon. It will be long and probably lonely as well. Will try to meet people on the way. Will try to fall in love with the programming again.
All I know is, this is an idea whose time has come. This is an idea worth solving. It will be a huge computer science problem. Not just the recognition of plate but even the use cases we will be building around it. Looking at thousands of vehicles, storing their numbers on the server and acting on them, in real time or offline!
Right now, the immediate applications on the mind include:
1) Access/Security management: Identify vehicles coming and going out of an organisation. Store the data!
2) Parking: Not as a solution but to get the data.
Lots of things have happened since then, but not much related to A.Eye!
In the last few couple of weeks, I resumed working on A.Eye. Here are the developments.
1) There is a service by name OpenALPR, who has API for number plate detection. It's decently good. Our approach will be to start using this till we have in-house solution!
2) Srinivas Anna won't be fully part of this company. He is relocating to Microsoft US. That is hugely unfortunate for the company. It's kind of removing strong foundation of the company. Now, I will be required to learn things on do it on our own.
3) I have been doing some domain research in last few weeks and realised few things.
1) This is a pretty old field. As old as 1970 where it was used as a surveillance by the police.
2) It required huge setup to run ANPR.
3) There are multiple companies working in this field across different countries and also at home!
4) I haven't come across any big company who is solely working on this project and it's applications. I am sure there will be other startups in this domain, but none of them are yet fully into markets.
Running it alone won't be easy. There are too many things involved. Android, Hardware, AI, Marketing, Cameras and everything else involved in building a startup, including getting funded!
The goal here is to build a platform to get the highest accuracy in ANPR for vehicles in India. The goal is to read any number plate which is readable, with a normal low cost camera connected to internet.
Like any other startup founder, I have no idea why no one else has already done it. I mean, get the highest accuracy on the number plates. And build applications around it.
So, knowingly or unknowingly we are getting into this field called computer vision. I have no idea how far I am from mastering this domain like an expert. Or how much time it takes to do that.
The journey will be starting soon. It will be long and probably lonely as well. Will try to meet people on the way. Will try to fall in love with the programming again.
All I know is, this is an idea whose time has come. This is an idea worth solving. It will be a huge computer science problem. Not just the recognition of plate but even the use cases we will be building around it. Looking at thousands of vehicles, storing their numbers on the server and acting on them, in real time or offline!
Right now, the immediate applications on the mind include:
1) Access/Security management: Identify vehicles coming and going out of an organisation. Store the data!
2) Parking: Not as a solution but to get the data.
Comments
Post a Comment