R. Khan

Bio: R. Khan is an academic researcher. The author has contributed to research in topics: Bag-of-words model & Visual Word. The author has an hindex of 1, co-authored 2 publications receiving 6 citations.

Evaluation of sift and surf using bag of words model on a large dataset

Abstract: In this paper, the objective is image classification analysis based on the well known image descriptors, the Scale Invariant Feature Transform (SIFT) and the Speeded up Robust Features (SURF) on five online available standard datasets. For the classification framework, we adopted the visual words approach. For SIFT, we use the Lowe’s implementation and for Speeded up Robust Features (SURF), the Herbert Bay’s implementation is used. Extensive experimentation using five datasets shows that SURF is a better choice compared to Scale Invariant Feature Transform (SIFT).

On experimental comparison of color and supervised face detection

Abstract: Viola-Jones algorithm and skin detection are the two widely used approaches for face detection. This paper aims at showing, experimentally, the trade-offs associated with each of the algorithm. The parameters used in this work are the ones encountered in real images. These include the illumination, color-space representation, face orientation, image sharpness and image complexity. Experimentally, the peculiarities associated with each algorithm are shown. Based on these experimental results, the situations are commented upon where these algorithms will be efficacious. This paper moreover aims at to provide an objective comparison of the two algorithms rather than going into the subjective details of peculiarities of any individual implementation

Towards an All-Purpose Content-Based Multimedia Information Retrieval System

Abstract: The growth of multimedia collections - in terms of size, heterogeneity, and variety of media types - necessitates systems that are able to conjointly deal with several forms of media, especially when it comes to searching for particular objects. However, existing retrieval systems are organized in silos and treat different media types separately. As a consequence, retrieval across media types is either not supported at all or subject to major limitations. In this paper, we present vitrivr, a content-based multimedia information retrieval stack. As opposed to the keyword search approach implemented by most media management systems, vitrivr makes direct use of the object's content to facilitate different types of similarity search, such as Query-by-Example or Query-by-Sketch, for and, most importantly, across different media types - namely, images, audio, videos, and 3D models. Furthermore, we introduce a new web-based user interface that enables easy-to-use, multimodal retrieval from and browsing in mixed media collections. The effectiveness of vitrivr is shown on the basis of a user study that involves different query and media types. To the best of our knowledge, the full vitrivr stack is unique in that it is the first multimedia retrieval system that seamlessly integrates support for four different types of media. As such, it paves the way towards an all-purpose, content-based multimedia information retrieval system.

Evaluation of Basic Convolutional Neural Network, AlexNet and Bag of Features for Indoor Object Recognition

Abstract: This paper evaluates two deep learning techniques that are basic Convolutional Neural Network (CNN) and AlexNet along with a classical local descriptor that is Bag of Features (BoF) with Speeded-Up Robust Feature (SURF) and Support Vector Machine (SVM) classifier for indoor object recognition. A publicly available dataset, MCIndoor20000, has been used in this experiment that consists of doors, signage, and stairs images of Marshfield Clinic. Experimental results indicate that AlexNet achieves the highest accuracy followed by basic CNN and BoF. Furthermore, the results also show that BoF, a machine learning technique, can also produce a high accuracy performance as basic CNN, a deep learning technique, for image recognition.

Evaluation of basic convolutional neural network and bag of features for leaf recognition

Abstract: This paper presents the evaluation of basic Convolutional Neural Network (CNN) and Bag of Features (BoF) for Leaf Recognition. In this study, the performance of basic CNN and BoF for leaf recognition using a publicly available dataset called Folio dataset has been investigated. CNN has proven its powerful feature representation power in computer vision. The same goes with BoF where it has set new performance standards on popular image classification benchmarks and has achieved scalability breakthrough in image retrieval. The feature that is being utilized in the BoF is Speeded-Up Robust Feature (SURF) texture feature. The experimental results indicate that BoF achieves better accuracy compared to basic CNN.

Convolutional neural network vs bag of features for bambara groundnut leaf disease recognition

Abstract: This paper investigates bambara groundnut leaf disease recognition using two popular techniques known as Convolutional Neural Network (CNN) and Bag of Features (BOF) with Speeded-up Robust Feature (SURF) and Support Vector Machine (SVM) classifier. Leaf disease recognition has attracted many researchers because the outcome is useful for farmers. One of the crops that provide high income for farmers is bambara groundnut but the leaves are easily infected with diseases especially after the rain. This could affect the crop productivity. Thus, automatic disease recognition is crucial. A new dataset that consists of 400 images of the infected and non-infected leaves of bambara groundnut has been constructed. The experimental results indicate that both of these techniques produce excellent leaf disease recognition accuracy.

Bag of Features vs Deep Neural Networks for Face Recognition

Abstract: This paper proposes a comparative study of Bag of Features (BoF) and Deep Neural Networks (DNN) approaches for the problem of face recognition. For the latter approach we consider three pre-trained models, namely AlexNet, ResNet50 and GoogleNet provided through Caffe Model Zoo and use them as feature extractors. Although these models were trained on different datasets, e.g., ImageNet, bottom-most layers act like universal feature extractors thus it is possible to be employed for different classification tasks. In order to adapt the models to various face datasets requirements we performed modifications to the input data as well as to the output layer of the pre-trained models by replacing it with a multiclass SVM classifier.