ADAPTIVE INDOOR SCENE CLASSIFICATION WITH MULTI-SVM CLASSIFICATION TO SOLVE MULTI-CLASS PROBLEM
Journal Title: International Journal of Engineering Sciences & Research Technology - Year 30, Vol 5, Issue 5
Abstract
In recent years, indoor scene recognition has attracted much attention and its research has rapidly expanded by not just engineers but also neuroscientists, since it has many potential applications in computer vision communication and automatic access control system. Especially, indoor scene recognition is an important part of computer vision and feature recognition as the first step of automatic uninterrupted robotic movement or computer vision applications like automatic interior designing algorithms. However, the indoor scene detection is not straightforward because it has lots of variations of image appearance, such as light effect, occlusion, image orientation, illuminating condition and object variety. Many novel methods have been advanced to resolve each variation listed above. For example, the templatematching methods are used for indoor scene localization and detection by computing the correlation of an input image to a standard and training scene appearance or pattern. The feature invariant approaches are used for feature detection of bed, chair, cabinet, table, door, electrical or electronic items, etc. The appearance-based methods are used for indoor feature detection with support vector machine and information theoretical approach. Nevertheless, implementing the methods altogether is still a difficult task. Fortunately, the images used in this project have some degree of uniformity thus the detection algorithm can be easy: first, the all the faces are vertical and have frontal view; second, they are under almost the same illuminate state. This project presents an indoor scene detection technique mainly based on the appearance based feature segmentation, SVM training and SVM classification methods to recognize the indoor scenes.
Authors and Affiliations
Japneet Kaur
Keywords
Related Articles
- EP ID EP154155
- DOI 10.5281/zenodo.52502
- Views 59
- Downloads 0
Improving Performance of KNN Algorithm by Using Enhanced CBIR Techniques for Retrieval of MRI Images
CBIR (Content Based Image Retrieval) is a method utilized for the recovery of coveted pictures from a huge collection of pictures in a database on the premise of picture characteristics, for example, coloration, s...
Robustness of PD and PID Controllers used with Second-Order Processes
Robustness is one of the requirements used in controllers and compensators design. The designs presented in the previous papers did not consider the robustness of the controller or compensator. Therefore, the obje...
BRAINY MOTE- AN AREA EFFICIENT SMART SENSOR
The new technology is the INTERNET OF THINGS, the general concept of the Internet of Things is that we can put a sensor on anything and have it send data back to a database through the Internet. In this way we can...
Nanotechnology in Green Chemistry- A study
Nanotechnologies as well as Nano scale technologies refer to the broad range of research and applications whose common trait is size. Nanotechnology may be able to create many new materials and devices with a vast...
UPQC Control of Wind Turbine Fed FSIG under Asymmetric Faults
his paper presents the mitigation of faults in w unified power quality conditioner because of its excellent performance of fault mitigation. The UPQC consists of shunt and series converters connected back dc ste...