Sensors to detect attendance of students in class.

This was very interesting project which I had thought of making in my college days years ago. There is no sensor to detect particular person as likely to work as biometrics . I thought to do tweaks with some PIR motion sensor but it cannot count.

Then someone suggested me to use the person's cellphone as their identity and the simple ceiling detector is able to detect a particular code/message from cell no. and it accounts to the student attendance but again this is somewhat not practically feasible. there are many loopholes and constraints to this.

Next possible thing came to my mind was to implement a system which could monitor the audience through their respective sitting places. particular seats will be allotted to the students and the system just has to monitor the vacant/occupied and will mark respective student's name. if the seat will remain vacant, absent will be marked in front of student's name. I am trying a lot to implement the prototype even but not much successful.

You just think of reality shows going on where I have to take audience poll, they have respective meters in their hand and system is able to get voting poll, can we make some tweaks to this system and  convert it into meaning full attendance system.... Can we do something out of box??

Sensors can detect a specific person if that person is uniquely tagged, perhaps with an RFID tag, for example.  However, a tagging system can be defeated if the tag is not carried consitently, if the tag is shielded from RF, or if someone else carries the tag in a purposeful attempt to game the system.  Biometric identification based on facial recognition is getting much better, but is still expensive and the ethical issues are huge and complex.  A low tech solution is to have students hand in signed work at the end of every class.  If the work is derived from topics discussed during the class, then spoofing is pretty difficult.

I think the best approach to managing attendance is to make sure classes are engaging, relevant, and respectful of the needs of the learner.  If those aspects are addressed then worrying about attendance should be pretty much a non-issue.

RFID tags can be useful if they can be injected into human body to work like a unique identification mark,but system is too costly. Microsoft uses the same phenomenon to track their employees position , outcome will be foolproof to the system.

As other users has already answered, sensors can detect a specific unit (person, animal or whatever else) if the unit has a unique ID. At the actual date the most reliable and widely diffused system is just the RFID but the detection range is short and this kind of solutions are used to detect someone just when passing in specific points (i.e. parkind, entry / exit gate etc.).

The most (probably illegal) reliable way to track a person instead is following - using different kind of sensors / RF detection systems - through its mobile device or a GPS unit.

Enrico

Sir,

          I have studied about RFID technique,but there are some loopholes in this process.I think that if a person is not interested to come to college he/she can send his/her id card with his friend so he/she can get attendance.Even face detection is also not possible if there are twins in the class.So i need a better technique for this.

When I was at school we had this new thing called a class register.

Instructorman has already made a really good point about engaging with the class.

I think the best approach to managing attendance is to make sure classes are engaging, relevant, and respectful of the needs of the learner.

It seems like this technology is trying to solve a problem that doesn't exist.

In my experience some schools in the UK are unwilling to implement electronic methoulds of attendance detection due to the risk of losing data. A local school recently received a hefty fine after losing data from the electronic registration system. I agree with Paul that physical registers are enough to solve the problem.

I'm sure this is  a case where traditional technique works best. At the end of the year you have exams which are conducted under carefully controlled conditions. If the student passes does it matter if he learned by reading, attending lectures or hanging out in bars ? (NB this last approach is not usually successful)

MK

The thing, I suggest is to wore a batch,to sense a persons presence inside the class.

the batch must be provided with a code,(the code refers to the person's personal identification which is physical) .