I am trying to learn how Neural Network works on image recognition.I am confused that how neural network that how i will give input.My definition is find(track) object in sequence of images(particular that image come from aircraft travelling & capturing images from camera.so in that most of objects are same so how to track this type of object with neural network) .What input should given to neural network?
In generally, we use neural networks only for recognize the content of images. Tracking has two different usage; (1). tracking an object through a set of frames grabbed from a video (2). Re-recognizing an object uniquely when ever it appeared in another place.
The first application requires one time detection of the object and the tracking can be done by using a flow analysis. The 2nd application is harder because we need to look into fine details of the object. At a glance an object may appeared to similar but in reality they are not.
Neural network can be used to recognize/detect object category but it will require more works to uniquely identify an object.
A classical neural network requires to input a set of features extracted from each of the image. Deep neural network (DNN) works with image pixels.
If you can elaborate the specific application type(from which I mentioned above), I may be able to suggest a precise method.
https://arxiv.org/abs/1612.08242
This article and method might be useful in understanding it more in depth. You might find some videos on YouTube based on this algorithm where they show object recognition in video feed.
Neural networks in my view have two major problems. First they require large amounts of training data which do not necessarily span the search space. If there is supporting theory which guarantees this, then that is fine (eg. chess problems). Secondly they require measurements to be selected to drive the network. There is no guarantee that these measurements will be present in unseen data. For example in visual data, the lighting changes, part of the object is occluded, the object is distorted, etc. Other approaches are needed.
To analyze the movement of objects in the image, hopefully this will be useful for you. "Object Detection and Tracking for Mobile Image Monitoring Systems"
Conference Paper Deteksi dan Tracking Objek untuk Sistem Pengawasan Citra Bergerak
- Badges
- Science topic
- Similar topics
- Computer Science
- Images