Siamese Network Object Tracking Algorithm Combining Attention Mechanism and Correlation Filter Theory

Aiming to solve the problem of tracking drift during movement, which was caused by the lack of discriminability of the feature information and the failure of a fixed template to adapt to the change of object appearance, the paper proposes an object tracking algorithm combining attention mechanism and correlation filter theory based on the framework of full convolutional Siamese neural networks. Firstly, the apparent information is processed by using the attention mechanism thought, where the object and search area features are optimized according to the spatial attention and channel attention module. At the same time, the cross-attention module is introduced to process the template branch and search area branch, respectively, which makes full use of the diversified context information of the search area. Then, the background perception correlation filter model with scale adaptation and learning rate adjustment is adopted into the model construction, using as a layer in the network model to realize the object template update. Finally, the optimal object location is determined according to the confidence map with similarity calculation. Experimental results show that the designed method in the paper can promote the object tracking performance under various challenging environments effectively; the success rate increases by 16.2%, and the accuracy rate increases by 16%.

Fast Algorithm for Calculating Transitive Closures of Binary Relations in the Structure of a Network Object

The article proposes a fast algorithm for constructing the transitive closures between all pairs of nodes in the structure of a network object, which can have both directional and non-directional links. The algorithm is based on the disjunctive addition of the elements of certain rows of the adjacency matrix, which models (describe) the structure of the original network object. The article formulates and proves a theorem that using such a procedure, the matrix of transitive closures of a network object can be obtained from the adjacency matrix in two iterations (traversal) on such an array. An estimate of the asymptotic computational complexity of the proposed algorithm is substantiated. The article presents the results of an experimental study of the execution time of such an algorithm on network structures of different dimensions and with different connection densities. For this indicator, the developed algorithm is compared with the well-known approaches of Bellman, Warshall-Floyd, Shimbel, which can also be used to determine the transitive closures of binary relations of network objects. The corresponding graphs of the obtained dependences are given. The proposed algorithm (the logic embedded in it) can become the basis for solving problems of monitoring the connectivity of various subscribers in data transmission networks in real time when managing the load in such networks, where the time spent on routing information flows directly depends on the execution time of control algorithms, as well as when solving other problems on the network structures.

Congratulations ! JACIII Best Paper and Young Researcher Awards 2021

We are pleased to announce that the JACIII Awards of 2021 have been decided by the JACIII editorial boards. This year, the award winning papers were severely and fairly selected among 362 papers published in JACIII Vols. 22 (2018) to 24 (2020) and there was no entries that deserved the Best Review Paper award. The award ceremony was held online in order to prevent spreading of COVID-19. JACIII BEST PAPER AWARD 2021 Sotetsu Suzugamine, Takeru Aoki, Keiki Takadama, and Hiroyuki Sato Self-Structured Cortical Learning Algorithm by Dynamically Adjusting Columns and Cells JACIII Vol.24 No.2, pp. 185-198, 2020. JACIII YOUNG RESEARCHER AWARD 2021 JACIII YOUNG RESEARCHER AWARD 2021 Xiaobo Liu Jinxin Chi Emotion Recognition Based on Multi-Composition Deep Forest and Transferred Convolutional Neural Network Object-Oriented 3D Semantic Mapping Based on Instance Segmentation By Xiaobo Liu, Xu Yin, Min Wang, Yaoming Cai, and Guang Qi By Jinxin Chi, Hao Wu, and Guohui Tian JACIII Vol.23 No.5, pp. 883-890, 2019. JACIII Vol.23 No.4, pp. 695-704, 2019.

University Network Capital as an Element of the City’s Social Capital

The formation and development of network capital in the university brings up the issue of its influence on characteristics of the city’s social capital. The basic provisions of the network approach and the theory of social capital are used as theoretical grounds to identify such an impact. The article reveals the features of a modern university as a complex network object and the characteristics of interpersonal relationships that arise in its socio-cultural environment. These include the cultural homogeneity, formation of a joint reality, optimization of communication processes, the need for cooperation. Taking into account the revealed specificity, the network capital available to the subjects of social interaction is characterized and studied through the categories of trust and solidarity. It is determined that trust and solidarity are not only central, but necessary structural elements of the university’s network capital, first of all, because the peculiarities of the organization of network interaction dictate the preferred strategies of network behavior, into which these phenomena are embedded. Thus, being the holder of intellectual resources and a network of interpersonal connections, the university produces the effective ways to expand the city’s social capital, firstly, through a unique social network organization capable of rapid mobilization. It provides access to the formation of temporary teams with deep and versatile competencies that generate “quick trust”. Secondly, it expands the city’s social capital through impersonal trust, which convinces society that the university has some universal competence related to the life of the city and its people.

Automatic Fruit Detection and Couting System Using Neural Network

In the field of agriculture, Identification and counting the number of fruits from the image helps the farmers in crop estimation. At present manual counting of fruits present in many places. The current practice of yield estimation based on the manual counting of fruits has many drawbacks as it is time consuming and expensive process. while considering the progress of fruit detection, estimating proper and accurate fruit counts from images in real-world scenarios such as orchards is still a challenging problem. The focus of this paper is on the web application of fruit yield estimation. This web application helps the farmers to count the number of fruits easily. This system provides an automated and efficient fruit counting system using computer vision techniques. This paper provides the progress towards in-field fruit counting using neural network object detection methods. So this process is done by recognizing each fruit in the image and taking the count. In the neural network, we have used YOLO architecture for recognizing the fruits.

A novel method for finding grasping handles in a clutter using RGBD Gaussian mixture models

The paper proposes a novel method to detect graspable handles for picking objects from a confined and cluttered space, such as the bins of a rack in a retail warehouse. The proposed method combines color and depth curvature information to create a Gaussian mixture model that can segment the target object from its background and imposes the geometrical constraints of a two-finger gripper to localize the graspable regions. This helps in overcoming the limitations of a poorly trained deep network object detector and provides a simple and efficient method for grasp pose detection that does not require a priori knowledge about object geometry and can be implemented online with near real-time performance. The efficacy of the proposed approach is demonstrated through simulation as well as real-world experiment.