scholarly journals Inferring Visual Biases in UAV Videos from Eye Movements

Drones ◽  
2020 ◽  
Vol 4 (3) ◽  
pp. 31
Author(s):  
Anne-Flore Perrin ◽  
Lu Zhang ◽  
Olivier Le Meur

Unmanned Aerial Vehicle (UAV) imagery is gaining a lot of momentum lately. Indeed, gathered information from a bird-point-of-view is particularly relevant for numerous applications, from agriculture to surveillance services. We herewith study visual saliency to verify whether there are tangible differences between this imagery and more conventional contents. We first describe typical and UAV contents based on their human saliency maps in a high-dimensional space, encompassing saliency map statistics, distribution characteristics, and other specifically designed features. Thanks to a large amount of eye tracking data collected on UAV, we stress the differences between typical and UAV videos, but more importantly within UAV sequences. We then designed a process to extract new visual attention biases in the UAV imagery, leading to the definition of a new dictionary of visual biases. We then conduct a benchmark on two different datasets, whose results confirm that the 20 defined biases are relevant as a low-complexity saliency prediction system.

This chapter is an illustration of feature extraction for working with large datasets. The basic definition of feature extraction, selection of effective features, and the existing problems and solutions are provided. How feature extraction maps the high dimensional space to smaller space is explained.


Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-17
Author(s):  
Yildiz Aydin ◽  
Bekir Dizdaroğlu

Degradations frequently occur in archive films that symbolize the historical and cultural heritage of a nation. In this study, the problem of detection blotches commonly encountered in archive films is handled. Here, a block-based blotch detection method is proposed based on a visual saliency map. The visual saliency map reveals prominent areas in an input frame and thus enables more accurate results in the blotch detection. A simple and effective visual saliency map method is taken into consideration in order to reduce computational complexity for the detection phase. After the visual saliency maps of the given frames are obtained, blotch regions are estimated by considered spatiotemporal patches—without the requirement for motion estimation—around the saliency pixels, which are subjected to a prethresholding process. Experimental results show that the proposed block-based blotch detection method provides a significant advantage with reducing false alarm rates over HOG feature (Yous and Serir, 2017), LBP feature (Yous and Serir, 2017), and regions-matching (Yous and Serir, 2016) methods presented in recent years.


2011 ◽  
Vol 54 (1) ◽  
pp. 161-186 ◽  
Author(s):  
Liam O'Carroll ◽  
Francesc Planas-Vilanova

AbstractThis paper takes a new look at ideals generated by 2×2 minors of 2×3 matrices whose entries are powers of three elements not necessarily forming a regular sequence. A special case of this is the ideals determining monomial curves in three-dimensional space, which were studied by Herzog. In the broader context studied here, these ideals are identified as Northcott ideals in the sense of Vasconcelos, and so their liaison properties are displayed. It is shown that they are set-theoretically complete intersections, revisiting the work of Bresinsky and of Valla. Even when the three elements are taken to be variables in a polynomial ring in three variables over a field, this point of view gives a larger class of ideals than just the defining ideals of monomial curves. We then characterize when the ideals in this larger class are prime, we show that they are usually radical and, using the theory of multiplicities, we give upper bounds on the number of their minimal prime ideals, one of these primes being a uniquely determined prime ideal of definition of a monomial curve. Finally, we provide examples of characteristic-dependent minimal prime and primary structures for these ideals.


2011 ◽  
Vol 63-64 ◽  
pp. 350-354 ◽  
Author(s):  
Li Li Lin ◽  
Neng Rong Chen

The background modeling method based on the Gaussian mixture model (GMM) is usually used to detect the moving objects in static background. But when applied to dynamic background, for example caused by camera jitter, the wrong detection rate of moving objects is high, and thus affects the follow-up tracking. In addition, the method with GMM can not effectively remove the moving objects shadow region. This paper proposes a moving object detection method based on GMM and visual saliency maps, which not only can remove the disturbance caused by camera jitter, but also can effectively solve the shadow problem and achieve stable moving objects detection.


Author(s):  
Yanyu Xu ◽  
Nianyi Li ◽  
Junru Wu ◽  
Jingyi Yu ◽  
Shenghua Gao

Saliency detection is a long standing problem in computer vision. Tremendous efforts have been focused on exploring a universal saliency model across users despite their differences in gender, race, age, etc. Yet recent psychology studies suggest that saliency is highly specific than universal: individuals exhibit heterogeneous gaze patterns when viewing an identical scene containing multiple salient objects. In this paper, we first show that such heterogeneity is common and critical for reliable saliency prediction. Our study also produces the first database of personalized saliency maps (PSMs). We model PSM based on universal saliency map (USM) shared by different participants and adopt a multi-task CNN framework to estimate the discrepancy between PSM and USM. Comprehensive experiments demonstrate that our new PSM model and prediction scheme are effective and reliable.


2020 ◽  
Vol 10 (23) ◽  
pp. 8335
Author(s):  
Yingtong Lu ◽  
Yaofei Ma ◽  
Jiangyun Wang ◽  
Liang Han

To perform air missions with an unmanned aerial vehicle (UAV) swarm is a significant trend in warfare. The task assignment among the UAV swarm is one of the key issues in such missions. This paper proposes PSO-GA-DWPA (discrete wolf pack algorithm with the principles of particle swarm optimization and genetic algorithm) to solve the task assignment of a UAV swarm with fast convergence speed. The PSO-GA-DWPA is confirmed with three different ground-attack scenarios by experiments. The comparative results show that the improved algorithm not only converges faster than the original WPA and PSO, but it also exhibits excellent search quality in high-dimensional space.


Author(s):  
Wei Xiong ◽  
Yongli Xu ◽  
Yafei Lv ◽  
Libo Yao

Targets detection in synthetic aperture radar (SAR) remote sensing images, which is a fundamental but challenging problem in the field of satellite image analysis, plays an important role for a wide range of applications and is receiving significant attention in recent years. Besides, the ability of human visual system to detect visual saliency is extraordinarily fast and reliable. However, computational modeling of SAR image scene still remains a challenge. This paper analyzes the defects and shortcomings of traditional visual models applied to SAR images. Then a visual attention model designed for SAR images is proposed. The model draws the basic framework of classical ITTI model; selects and extracts the texture features and other features that can describe the SAR image better. We proposes a new algorithm for computing the local texture saliency of the input image, then the model constructs the corresponding saliency maps of features; Next, a new mechanism of feature fusion is adopted to replace the linear additive mechanism of classical models to obtain the overall saliency map; Finally, the gray-scale characteristics of focus of attention (FOA) in saliency map of all features are taken into account, our model choose the best saliency representation, Through the multi-scale competition strategy, the filter and threshold segmentation of the saliency maps can be used to select the salient regions accurately, thereby completing this operation for the visual saliency detection in SAR images. In the paper, several types of satellite image data, such as TerraSAR-X (TS-X), Radarsat-2, are used to evaluate the performance of visual models. The results show that our model provides superior performance compared with classical visual models. By further contrasting with the classical visual models, Our model reduce the false alarm caused by speckle noise, and its detection speed is greatly improved, and it is increased by 25% to 45%.


2019 ◽  
Vol 30 (2) ◽  
pp. 109-122
Author(s):  
Aleksandar Bulajić ◽  
Miomir Despotović ◽  
Thomas Lachmann

Abstract. The article discusses the emergence of a functional literacy construct and the rediscovery of illiteracy in industrialized countries during the second half of the 20th century. It offers a short explanation of how the construct evolved over time. In addition, it explores how functional (il)literacy is conceived differently by research discourses of cognitive and neural studies, on the one hand, and by prescriptive and normative international policy documents and adult education, on the other hand. Furthermore, it analyses how literacy skills surveys such as the Level One Study (leo.) or the PIAAC may help to bridge the gap between cognitive and more practical and educational approaches to literacy, the goal being to place the functional illiteracy (FI) construct within its existing scale levels. It also sheds more light on the way in which FI can be perceived in terms of different cognitive processes and underlying components of reading. By building on the previous work of other authors and previous definitions, the article brings together different views of FI and offers a perspective for a needed operational definition of the concept, which would be an appropriate reference point for future educational, political, and scientific utilization.


2018 ◽  
Vol 6 (3) ◽  
Author(s):  
Wilson Otto Gomes Batista ◽  
Alexandre Gomes De Carvalho

Contrast-detail (C-D) curves are useful in evaluating the radiographic image quality in a global way. The objective of the present study was to obtain the C-D curves and the inverse image quality figure. Both of these parameters were used as an evaluation tool for abdominal and chest imaging protocols. The C-D curves were obtained with the phantom CDRAD 2.0 in computerized radiography and the direct radiography systems (including portable devices). The protocols were 90 and 102 kV in the range of 2 to 20 mAs for the chest and 80 kV in the range of 10 to 80 mAs for the abdomen. The incident air kerma values were evaluated with a solid state sensor. The analysis of these C-D curves help to identify which technique would allow a lower value of the entrance surface air kerma, Ke, while maintaining the image quality from the point of view of C-D detectability. The results showed that the inverse image quality figure, IQFinv, varied little throughout the range of mAs, while the value of Ke varied linearly directly with the mAs values. Also, the complete analysis of the curves indicated that there was an increase in the definition of the details with increasing mAs. It can be concluded that, in the transition phase for the use of the new receptors, it is necessary to evaluate and adjust the practised protocols to ensure, at a minimum, the same levels of the image quality, taking into account the aspects of the radiation protection of the patient.


Author(s):  
Olena Karpenko ◽  
Tetiana Stoianova

The article is devoted to the study of personal names from a cognitive point of view. The study is based on the cognitive concept that speech actually exists not in the speech, not in linguistic writings and dictionaries, but in consciousness, in the mental lexicon, in the language of the brain. The conditions for identifying personal names can encompass not only the context, encyclopedias, and reference books, but also the sound form of the word. In the communicative process, during a free associative experiment, which included a name and a recipient’s mental lexicon. The recipient was assigned a task to quickly give some association to the name. The aggregate of a certain number of reactions of different recipients forms the associative field of a proper name. The associative experiment creates the best conditions for identifying the lexeme. The definition of a monosemantic personal name primarily includes the search of what it denotes, while during the process of identifying a polysemantic personal name recipients tend have different reactions. Scientific value is posed by the effect of the choice of letters for the name, sound symbolism, etc. The following belong to the generalized forms of identification: usage of a hyperonym; synonyms and periphrases or simple descriptions; associations denoting the whole (name stimulus) by reference to its part (associatives); cognitive structures such as “stimulus — association” and “whole (stimulus) — part (associative)”; lack of adjacency; mysterious associations. The topicality of the study is determined by its perspective to identify the directions of associative identification of proper names, which is one of the branches of cognitive onomastics. The purpose of the study is to identify, review, and highlight the directions of associative identification of proper names; the object of the research is the names in their entirety and variety; its subject is the existence of names in the mental lexicon, which determines the need for singling out the directions for the associative identification of the personal names.


Sign in / Sign up

Export Citation Format

Share Document