Indoor Outdoor Scene Classification in Digital Images
نویسنده
چکیده
In this paper, we present a method to classify real-world digital images into indoor and outdoor scenes. Indoor class consists of four groups: bedroom, kitchen, laboratory and library. Outdoor class consists of four groups: landscape, roads, buildings and garden. Application considers real-time system and has a dedicated data-set. Input images are pre-processed and converted into gray-scale and is re-sized to “128x128” dimensions. Pre-processed images are sent to “Gabor filters”, which pre-computes filter transfer functions, which are performed on Fourier domain. The processed signal is finally sent to GIST feature extraction and the images are classified using “kNN classifier”. Most of the techniques have been based on the use of texture and color space features. As of date, we have been able to achieve 80% accuracy with respect to image classification. Keywords—Indoor, Outdoor, Image Classification, Scene.
منابع مشابه
Border ownership assignment in real images.
We explored whether spectral based cues, such as extremal edges, are useful for border ownership classification in real images, and we analyzed how different environments: indoor vs. outdoor affect the prediction. Our algorithmic approach is based on a random forest classifier using spectral and Gestalt-like texture grouping features. The classifier detects in real-time the points in the image,...
متن کاملDepth map generation by image classification
This paper presents a novel and fully automatic technique to estimate depth information from a single input image. The proposed method is based on a new image classification technique able to classify digital images (also in Bayer pattern format) as indoor, outdoor with geometric elements or outdoor without geometric elements. Using the information collected in the classification step a suitabl...
متن کاملExtraction of Attributes, Nature and Context of Images
In this paper a framework for deriving highlevel scene attributes from low-level image features is presented. Examples of attributes derived are photo-painting, indoor-outdoor, night-day and nature-city. The assignment of the attributes to images is done by a hierarchical classification of the low level features, which capture colour, texture and spatial information. A prototype for image class...
متن کاملIndoor vs outdoor classification of consumer photographs using low-level and semantic features
Scene categorization to indoor vs outdoor may be approached by using low-level features for inferring high-level information about the image. Low-level features such as color and texture have been used extensively in image understanding research, however, they cannot solve the problem completely. In this paper, we propose the use of a Bayesian network for integrating knowledge from low-level an...
متن کاملText Detection in Indoor/Outdoor Scene Images
In this paper, we propose a novel methodology for text detection in indoor/outdoor scene images. The proposed methodology is based on an efficient binarization and enhancement technique followed by a suitable connected component analysis procedure. Image binarization successfully process indoor/ outdoor scene images having shadows, non-uniform illumination, low contrast and large signal-depende...
متن کامل