MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

Object detectors are typically trained on a large set of still images annotated by bounding-boxes. This paper introduces an approach for learning object detectors from real-world web videos known only to contain objects of a target class. We propose a fully automatic pipeline that localizes objects in a set of videos of the class and learns a detector for it. The approach extracts candidate spatio-temporal tubes based on motion segmentation and then selects one tube per video jointly over all videos. To compare to the state of the art, we test our detector on still images, i.e., Pascal VOC 2007. We observe that frames extracted from web videos can differ significantly in terms of quality to still images taken by a good camera. Thus, we formulate the learning from videos as a domain adaptation task. We show that training from a combination of weakly annotated videos and fully annotated still images using domain adaptation improves the performance of a detector trained from still images alone.

/pdf/learning-object-class-detectors-from-weakly-annotated-video-3ta868ym3s.pdf

Learning object class detectors from weakly annotated video

Vessel detection and classification from spaceborne optical images: A literature survey.

The objective of this work is to determine if people are interacting in TV video by detecting whether they are looking at each other or not. We determine both the temporal period of the interaction and also spatially localize the relevant people. We make the following four contributions: (i) head detection with implicit coarse pose information (front, profile, back); (ii) continuous head pose estimation in unconstrained scenarios (TV video) using Gaussian process regression; (iii) propose and evaluate several methods for assessing whether and when pairs of people are looking at each other in a video shot; and (iv) introduce new ground truth annotation for this task, extending the TV human interactions dataset (Patron-Perez et al. 2010) The performance of the methods is evaluated on this dataset, which consists of 300 video clips extracted from TV shows. Despite the variety and difficulty of this video material, our best method obtains an average precision of 87.6 % in a fully automatic manner.

/pdf/detecting-people-looking-at-each-other-in-videos-2txk5fsktl.pdf

Detecting People Looking at Each Other in Videos

Background Subtraction in Real Applications: Challenges, Current Models and Future Directions

This paper presents a method to accurately detect and monitor ships within the area of interest It is an advanced version of the previous works done regarding moving ship detection and tracking The proposed tracking scheme is based on the characteristics of both sea and ship, which includes: background information and local position of the ship Background subtraction and registration is achieved using morphological ‘Open’ operation and the ships are located using their edge information The experimental results demonstrate robust and real-time ship detection and tracking with 987% detection rate The proposed algorithm will be useful in coastal surveillance and monitoring applications

Multiple Ship Detection and Tracking Using Background Registration and Morphological Operations

In this paper, we propose an improved RC6 encryption algorithm that have the same structure of encryption and decryption. So far, conventional RC6 algorithms have difference structure of encryption and decryption. We devise our algorithm by inserting a symmetric layer using simple rotation and XOR operations, in which the half of whole RC6 rounds uses encryption procedure and the rest of it are employs decryption one. The symmetry layer is put between encryption part and decryption one. The proposed RC6 algorithm has the almost same speed compared with the conventional RC6 one. Nevertheless, the proposed algorithm improves encryption security by inserting the symmetric layer because a differential and linear analysis has a difficulty in analyzing an encrypted stream. The proposed algorithm will be useful to the applications which require the same procedure of encryption and decryption such as light mobile devices and RFIDs.

An improved RC6 algorithm with the same structure of encryption and decryption

In this paper, we propose an applicable technique in the real time applications using FPGA chip implementation technique which can be implemented by Haar wavelet based real time video watermarking for robust watermarking for copyright protection of video contents. Proposed method does algorithm design which can reduce computation complexity for hardware implantation and then implement hardware. In implementation result, we can confirm correct watermark inserting by FPGA simulation tool and cannot recognize concealed watermark in video sequence which insert watermark in original video sequence using implemented hardware watermark embedder system. Our implemented watermark embedding chipcan be useful in movie production and broadcasting companies that requires real-time based copyright protection system.

Implementation of Real Time Video Watermark Embedder Based on Haar Wavelet Transform Using FPGA

Shot Change Detection is a main technique for automatic temporal segmentation of video, which requires adaptive adjustment of thresholds. In this paper, we propose an adaptive shot change detection algorithm using histograms of frames within extension sliding window. The proposed method generates an adaptive threshold which is calculated by an average of absolute difference histogram within extension sliding window. We obtained better the detection rate than conventional methods maximally 10% in F1. The proposed method will be helpful in searching video data on various applications.

An adaptive Shot Change Detection algorithm using an average of absolute difference histogram within extension sliding window

Moving ship detection and tracking is the key technology in the intelligence coastal surveillance system. In this paper we present an adaptive and improved method to accurately detect and monitor ships within the area of interest. It is an advanced version of the previous works done regarding moving ship detection and tracking. The proposed tracking scheme is based on the characteristics of both sea and ship, which include: background information and local position of the ship. We used Morphological "open" operation and frame-subtraction method for background subtraction. The ships are located using the edge information along morphological "thicken", "dilate" and "bridge" operations. The experimental results demonstrate robust and real-time ship detection and tracking over thousands of image frames, and robustness against cluttered background. The detection rate achieved was 98.4%.

Jong-Nam Kim

Papers

Multiple Ship Detection and Tracking Using Background Registration and Morphological Operations

An improved RC6 algorithm with the same structure of encryption and decryption

Implementation of Real Time Video Watermark Embedder Based on Haar Wavelet Transform Using FPGA

An adaptive Shot Change Detection algorithm using an average of absolute difference histogram within extension sliding window

An adaptive moving ship detection and tracking based on edge information and morphological operations