|International Journal of Computer Applications
|Foundation of Computer Science (FCS), NY, USA
|Volume 180 - Number 40
|Year of Publication: 2018
|Authors: GaganDeep Singh, Sonika Jindal
GaganDeep Singh, Sonika Jindal . A Novel Classification Approach Capable Indoor Scene Picture Identification with Hybrid Feature Selection Algorithm. International Journal of Computer Applications. 180, 40 ( May 2018), 35-38. DOI=10.5120/ijca2018917055
Image scene classification is an integral part of several aspects of image process. Indoor and outside classification could be a elementary part of scene process because it is that the place to begin of the many linguistics scene analysis approaches. Several novel techniques are developed to tackle this drawback, however every technique depends on its own information of pictures therefore reducing the boldness within the success of every technique. The planned model is formed capable of operating with the variations within the indoor scene image dataset, that are noticed within the sort of the color, texture, light, image orientation, occlusion and color illuminations. many experiments has been conducted over the projected model for the performance analysis of the indoor scene recognition system within the planned model. The results of the proposed model are obtained in the type of the various performance parameters of applied mathematics errors, precision, recall, F1-measure and overall accuracy. The planned technique has clearly outperformed the present models within the terms of the accuracy. The planned model improvement has been recorded above ten percent for all of the evaluated parameters against the prevailing models based mostly upon SURF, FREAK, etc.