Dubravko Culibrk

Neural Network Approach to Background Modeling for Video Object Segmentation

This paper presents a novel background modeling and subtraction approach for video object segmentation. A neural network (NN) architecture is proposed to form an unsupervised Bayesian classifier for this application domain. The constructed classifier efficiently handles the segmentation in natural-scene sequences with complex background motion and changes in illumination. The weights of the proposed NN serve as a model of the background and are temporally updated to reflect the observed statistics of background. The segmentation performance of the proposed NN is qualitatively and quantitatively examined and compared to two extant probabilistic object segmentation algorithms, based on a previously published test pool containing diverse surveillance-related sequences. The proposed algorithm is parallelized on a subpixel level and designed to enable efficient hardware implementation.

Salient Motion Features for Video Quality Assessment

Design of algorithms that are able to estimate video quality as perceived by human observers is of interest for a number of applications. Depending on the video content, the artifacts introduced by the coding process can be more or less pronounced and diversely affect the quality of videos, as estimated by humans. While it is well understood that motion affects both human attention and coding quality, this relationship has only recently started gaining attention among the research community, when video quality assessment (VQA) is concerned. In this paper, the effect of calculating several objective measure features, related to video coding artifacts, separately for salient motion and other regions of the frames of the sequence is examined. In addition, we propose a new scheme for quality assessment of coded video streams, which takes into account salient motion. Standardized procedure has been used to calculate the Mean Opinion Score (MOS), based on experiments conducted with a group of non-expert observers viewing standard definition (SD) sequences. MOS measurements were taken for nine different SD sequences, coded using MPEG-2 at five different bit-rates. Eighteen different published approaches related to measuring the amount of coding artifacts objectively on a single-frame basis were implemented. Additional features describing the intensity of salient motion in the frames, as well as the intensity of coding artifacts in the salient motion regions were proposed. Automatic feature selection was performed to determine the subset of features most correlated to video quality. The results show that salient-motion-related features enhance prediction and indicate that the presence of blocking effect artifacts and blurring in the salient regions and variance and intensity of temporal changes in non-salient regions influence the perceived video quality.

Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification

The latest generation of convolutional neural networks (CNNs) has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%.

New approaches to encryption and steganography for digital videos

In this work we propose a novel type of digital video encryption that has several advantages over other currently available digital video encryption schemes. We also present an extended classification of digital video encryption algorithms in order to clarify these advantages. We analyze both security and performance aspects of the proposed method, and show that the method is efficient and secure from a cryptographic point of view. Even though the method is currently feasible only for a certain class of video sequences and video codecs, the method is promising and future investigations might reveal its broader applicability. Finally, we extend our approach into a novel type of digital video steganography where it is possible to disguise a given video with another video.

K-means based segmentation for real-time zenithal people counting

The paper presents an efficient and reliable approach to automatic people segmentation, tracking and counting, designed for a system with an overhead mounted (zenithal) camera. Upon the initial block-wise background subtraction, k-means clustering is used to enable the segmentation of single persons in the scene. The number of people in the scene is estimated as the maximal number of clusters with acceptable inter-cluster separation. Tracking of segmented people is addressed as a problem of dynamic cluster assignment between two consecutive frames and it is solved in a greedy fashion. Systems for people counting are applied to people surveillance and management and lately within the ambient intelligence solutions. Experimental results suggest that the proposed method is able to achieve very good results in terms of counting accuracy and execution speed.

Dynamic saliency models and human attention: a comparative study on videos

Significant progress has been made in terms of computational models of bottom-up visual attention (saliency). However, efficient ways of comparing these models for still images remain an open research question. The problem is even more challenging when dealing with videos and dynamic saliency. The paper proposes a framework for dynamic-saliency model evaluation, based on a new database of diverse videos for which eye-tracking data has been collected. In addition, we present evaluation results obtained for 4 state-of-the-art dynamic-saliency models, two of which have not been verified on eye-tracking data before.

Digital video encryption algorithms based on correlation-preserving permutations

A novel encryption model for digital videos is presented. The model relies on the encryption-compression duality of certain types of permutations acting on video frames. In essence, the proposed encryption process preserves the spatial correlation and, as such, can be applied prior to the compression stage of a spatial-only video encoder. Several algorithmic modes of the proposed model targeted for different application requirements are presented and analyzed in terms of security and performance. Experimental results are generated for a number of standard benchmark sequences showing that the proposed method, in addition to providing confidentiality, preserves or improves the compression ratio.

A hybrid color-based foreground object detection method for automated marine surveillance

This paper proposes a hybrid foreground object detection method suitable for the marine surveillance applications. Our approach combines an existing foreground object detection method with an image color segmentation technique to improve accuracy. The foreground segmentation method employs a Bayesian decision framework, while the color segmentation part is graph-based and relies on the local variation of edges. We also establish the set of requirements any practical marine surveillance algorithm should fulfill, and show that our method conforms to these requirements. Experiments show good results in the domain of marine surveillance sequences.

Multiscale background modelling and segmentation

A new multiscale approach to motion based segmentation of objects in video sequences is presented. While image features extracted at multiple scales are commonly used within the pattern recognition community, they have seldom been employed for background modelling and subtraction. The paper describes a methodology for maintaining an explicit background model at multiple scales. Biological inspiration is used to contrive simple, yet effective mechanisms for feature extraction, incorporation of information across multiple scales and segmentation. Results of experiments conducted using sequences from the domain of traffic surveillance are presented in the paper. They suggest that the proposed method is able to achieve good segmentation results. In addition, the evaluated variant of a multiscale segmentation algorithm is far less computationally intensive, able to achieve processing of higher frame rates in real time and requires an order of magnitude less memory resources than the commonly-used approach compared against.

Feature Selection for Neural-Network Based No-Reference Video Quality Assessment

Design of algorithms that are able to estimate video quality as perceived by human observers is of interest for a number of applications. Depending on the video content, the artifacts introduced by the coding process can be more or less pronounced and diversely affect the quality of videos, as estimated by humans. In this paper we propose a new scheme for quality assessment of coded video streams, based on suitably chosen set of objective quality measures driven by human perception. Specifically, the relation of large number of objective measure features related to video coding artifacts is examined. Standardized procedure has been used to calculate the Mean Opinion Score (MOS), based on experiments conducted with a group of non-expert observers viewing SD sequences. MOS measurements were taken for nine different standard definition (SD) sequences, coded using MPEG-2 at five different bit-rates. Eighteen different published approaches for measuring the amount of coding artifacts objectively were implemented. The results obtained were used to design a novel no-reference MOS estimation algorithm using a multi-layer perceptron neural-network.