Efficient common objects localization based on deep hybrid Siamese network

2021 ◽  
pp. 1-10
Author(s):  
Mona M. Moussa ◽  
Rasha Shoitan ◽  
Mohamed S. Abdallah
Keyword(s):  
Distance Matrix ◽  
Multiple Objects ◽  
Siamese Network ◽  
Generation Stage ◽  
Pascal Voc ◽  
The Common ◽  
Conventional Methods ◽  
Weakly Supervised ◽  
Bounding Boxes ◽  
Two Stages

Finding the common objects in a set of images is considered one of the recent challenges in different computer vision tasks. Most of the conventional methods have proposed unsupervised and weakly supervised co-localization methods to find the common objects; however, these methods require producing a huge amount of region proposals. This paper tackles this problem by exploiting supervised learning benefits to localize the common object in a set of unlabeled images containing multiple objects or with no common objects. Two stages are proposed to localize the common objects: the candidate box generation stage and the matching and clustering stage. In the candidate box generation stage, the objects are localized and surrounded by the bounding boxes. The matching and clustering stage is applied on the generated bounding boxes and creates a distance matrix based on a trained Siamese network to reflect the matching percentage. Hierarchical clustering uses the generated distance matrix to find the common objects and create clusters for each one. The proposed method is trained on PASCAL VOC 2007 dataset; on the other hand, it is assessed by applying different experiments on PASCAL VOC 2007 6×2 and Object Discovery datasets, respectively. The results reveal that the proposed method outperforms the conventional methods by 8% to 40% in terms of corloc metric.

scholarly journals An Anchor-Free Siamese Network with Multi-Template Update for Object Tracking

Electronics ◽  
2021 ◽  
Vol 10 (9) ◽  
pp. 1067
Author(s):  
Tongtong Yuan ◽  
Wenzhu Yang ◽  
Qian Li ◽  
Yuxia Wang
Keyword(s):  
Object Tracking ◽  
Correlation Energy ◽  
Feature Maps ◽  
Siamese Network ◽  
Template Update ◽  
Free Network ◽  
Multiple Prediction ◽  
Bounding Boxes ◽  
High Level ◽  
Speed And Accuracy

Siamese trackers are widely used in various fields for their advantages of balancing speed and accuracy. Compared with the anchor-based method, the anchor-free-based approach can reach faster speeds without any drop in precision. Inspired by the Siamese network and anchor-free idea, an anchor-free Siamese network (AFSN) with multi-template updates for object tracking is proposed. To improve tracking performance, a dual-fusion method is adopted in which the multi-layer features and multiple prediction results are combined respectively. The low-level feature maps are concatenated with the high-level feature maps to make full use of both spatial and semantic information. To make the results as stable as possible, the final results are obtained by combining multiple prediction results. Aiming at the template update, a high-confidence multi-template update mechanism is used. The average peak to correlation energy is used to determine whether the template should be updated. We use the anchor-free network to implement object tracking in a per-pixel manner, which computes the object category and bounding boxes directly. Experimental results indicate that the average overlap and success rate of the proposed algorithm increase by about 5% and 10%, respectively, compared to the SiamRPN++ algorithm when running on the dataset of GOT-10k (Generic Object Tracking Benchmark).

scholarly journals ANALYTICAL SOLUTION OF THE FREQUENCY OF THE LOAD OSCILLATION AT AN ARBITRARY GIRDER NODE IN THE SYSTEM MAPLE

2019 ◽  
pp. 3-3
Author(s):  
Mikhail N. Kirsanov ◽  
Dmitriy V. Tinkov
Keyword(s):  
Analytical Solution ◽  
Arbitrary Number ◽  
Analytical Form ◽  
Labor Costs ◽  
Induction Method ◽  
Regular Type ◽  
Final Formula ◽  
The Common ◽  
Two Stages ◽  
Elastic Lattice

Introduction. We study the oscillations of a massive load on a planar statically definable symmetric truss of a regular type with parallel belts. Truss weight is not included. Free vertical oscillations are considered. The stiffness of the truss rods is assumed to be the same, the deformations are elastic. Lattice of the truss is double with descending braces and racks. New in the formulation and solution of the problem is the analytical form of the solution, which makes it possible in practice to easily evaluate the frequency characteristics of the structure depending on an arbitrary number of truss panels and the location of the load. Materials and methods. The operators and methods of the system of computer mathematics Maple are used. To determine the forces in the rods, the knotting method is used. The common terms of the sequence of coefficients of solutions for different numbers of panels are obtained from solving linear homogeneous recurrent equations of various order, obtained by special operators of the Maple system. Dependence on two arbitrary natural parameters is revealed in two stages. First, solutions for fixed load positions are found, then these solutions are summarized into one final formula for frequency. Results. By a series of individual solutions to the problem of load oscillation using the double induction method, it was possible to find common members of all sequences. The solution is polynomial in both natural parameters. Graphs constructed for particular cases, showed the adequacy of the approach. The discontinuous non-monotonic nature of the intermittent change depending on the number of truss panels and some other features of the solution are noted. Conclusions. It is shown that the induction method, previously applicable mainly to statics problems with one parameter (number of truss panels), is fully operational to the problems of the oscillations of system with two natural parameters. It should be noted that significant labor costs and a significant increase in the time symbolic transformations in such tasks

Processing of gold-antimony concentrates

2021 ◽  
pp. 59-65
Author(s):  
A. V. Markelov ◽  
K. M. Falin ◽  
V. A. Puchkina ◽  
A. N. Titova
Keyword(s):  
Selective Extraction ◽  
Leaching Rate ◽  
Pressure Oxidation ◽  
Selective Precipitation ◽  
Refractory Gold ◽  
Atmospheric Leaching ◽  
The Common ◽  
Two Stages ◽  
Special Process

This paper describes the results of a study that looked at processing of goldantimony concentrates with selective extraction of antimony and gold as commodities. The common global practice of processing antimony sulphide concentrates (20–30% Sb) is based on alkaline sulphide leaching followed by precipitation of metallic antimony by electrowinning. However, application of this technique to process sulphide concentrates that, apart from antimony, also contain gold, can be difficult as, together with antimony, up to 10–15% of gold can leach to the solution. It takes a special process during final refining of cathode antimony to recover that gold. This paper describes a process that involves two stages of atmospheric leaching of antimony. The gold that leached to the solution is precipitated with zinc after the first stage of antimony leaching. Together with atmospheric leach tailings, it then goes to the pressure oxidation unit. This process helps oxidize the rest of the sulphides and release refractory gold. The resultant cake is processed following a standard sorption cyanidation technique. The paper looks at the antimony leaching rate and the rate at which gold leaches to the solution during this process. The paper describes the results of selective precipitation of gold from gold-antimony solutions and highlights certain features of this process. A series of tests was conducted to test the techniques of pressure oxidation of atmospheric leach tailings and cyanidation of the residue. The paper also describes a process that was developed for processing of goldantimony concentrates and precipitation of antimony and gold. An antimony recovery exceeding 90–95% can be achieved when using this process. At the same time, the percent of dissolved gold can be reduced from 10–15 tо 1–3%.

scholarly journals YOLO-Based Simultaneous Target Detection and Classification in Automotive FMCW Radar Systems

Sensors ◽  
2020 ◽  
Vol 20 (10) ◽  
pp. 2897 ◽  
Author(s):  
Woosuk Kim ◽  
Hyunwoong Cho ◽  
Jongseok Kim ◽  
Byungkwan Kim ◽  
Seongwook Lee
Keyword(s):  
Spatial Clustering ◽  
Radar Data ◽  
Support Vector ◽  
Automotive Radar ◽  
Radar Sensor ◽  
Fmcw Radar ◽  
Radar Systems ◽  
Conventional Methods ◽  
Two Stages ◽  
Deep Learning Model

This paper proposes a method to simultaneously detect and classify objects by using a deep learning model, specifically you only look once (YOLO), with pre-processed automotive radar signals. In conventional methods, the detection and classification in automotive radar systems are conducted in two successive stages; however, in the proposed method, the two stages are combined into one. To verify the effectiveness of the proposed method, we applied it to the actual radar data measured using our automotive radar sensor. According to the results, our proposed method can simultaneously detect targets and classify them with over 90% accuracy. In addition, it shows better performance in terms of detection and classification, compared with conventional methods such as density-based spatial clustering of applications with noise or the support vector machine. Moreover, the proposed method especially exhibits better performance when detecting and classifying a vehicle with a long body.

A capsule-based collision detection approach of irregular objects in virtual maintenance

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Yuxue Jin ◽  
Jie Geng ◽  
Zhiyi He ◽  
Chuan Lv ◽  
Tingdi Zhao
Keyword(s):  
Collision Detection ◽  
Design Problems ◽  
Content Type ◽  
Detection Approach ◽  
Secondary Stage ◽  
Bounding Boxes ◽  
Two Stages ◽  
Maintenance Personnel ◽  
Virtual Maintenance ◽  
Encapsulation Methods

Purpose Virtual maintenance simulation is of great importance to help designers find and avoid design problems. During its simulation phase, besides the high precision requirement, collision detection must be suitable for all irregular objects in a virtual maintenance environment. Therefore, in this paper, a collision detection approach is proposed based on encapsulation for irregular objects in the virtual maintenance environment. Design/methodology/approach First, virtual maintenance simulation characteristics and several commonly used bounding boxes methods are analyzed, which motivates the application of encapsulation theory. Based on these, three different encapsulation methods are oriented to the needs of simulation, including encapsulation of rigid maintenance objects, flexible maintenance objects and maintenance personnel. In addition, to detecting collisions accurately, this paper divides the detection process into two stages. That is, in the first stage, a rough detection is carried out and then a tiny slice space is constructed to generate corresponding capsule groups, which will be redetected in the secondary stage. At last, several case studies are applied to illustrate the performance of the methodology. Findings The automatic construction algorithm for bounding boxes can be adapted to all forms of objects. The number of detection primitives are greatly reduced. It introduces the reachable space of the human body in maintainability as the collision search area. Originality/value The advantages of virtual maintenance simulation could also be advantageous in the industry with further studies. The paper believes this study is of particular interest to the readers of your journal.

scholarly journals Induction of chalcone synthase and phenylalanine ammonia-lyase by salicylic acid and Colletotrichum lindemuthianum in common bean

2003 ◽  
Vol 15 (3) ◽  
pp. 129-134 ◽  
Author(s):  
Ângela Diniz Campos ◽  
Alfredo Gui Ferreira ◽  
Magdolna Maria Vozári Hampe ◽  
Irajá Ferreira Antunes ◽  
Nely Brancão ◽  
...  
Keyword(s):  
Enzyme Activity ◽  
Salicylic Acid ◽  
Common Bean ◽  
Phenylalanine Ammonia Lyase ◽  
Chalcone Synthase ◽  
Colletotrichum Lindemuthianum ◽  
Water Control ◽  
Leaf Extracts ◽  
The Common ◽  
Two Stages

The activities of the enzymes chalcone synthase (CHS) and phenylalanine ammonia-lyase (PAL) were measured in leaf extracts obtained from four cultivars of the common bean (AB 136, Rio Tibagi, Carioca and Macanudo). Two stages of plant development were examined: plantlets (V2) and the onset of blooming (R6). Initially, the plants were either treated with salicylic acid or inoculated with the delta race of Colletotrichum lindemuthianum (inductive fungus) and after three days they were evaluated for enzyme activity. Afterwards, all plants were inoculated (challenged) with the virulent pathotype 33/95 of C. lindemuthianum except for the water control. Five days later, the activities of PAL and CHS were evaluated. There were significant changes in the activities of both enzymes three days after treatment with salicylic acid or inductive fungus when compared to the control. Five days after inoculation with with the virulent pathotype 33/95 of C. lindemuthianum CHS activity in the Macanudo was similar to control plants that were not treated with salicylic acid or the inductive fungus but inoculated with 33/95 C. lindemuthianum. The increase in enzyme activity after challenge with 33/95 C. lindemuthianum was greatest for the salicylic acid treatment in the cultivar AB 136, followed by Rio Tibagi and Carioca.

scholarly journals Stratified randomization for platform trials with differing experimental arm eligibility

2021 ◽  
pp. 174077452110288
Author(s):  
Subodh Selukar ◽  
Susanne May ◽  
Dave Law ◽  
Megan Othus
Keyword(s):  
Cancer Therapeutics ◽  
Standard Of Care ◽  
Worked Examples ◽  
Single Trial ◽  
Eligibility Criteria ◽  
Safety Issues ◽  
Use Of Resources ◽  
The Common ◽  
Conventional Methods ◽  
Do So

Background: Platform trials facilitate efficient use of resources by comparing multiple experimental agents to a common standard of care arm. They can accommodate a changing scientific paradigm within a single trial protocol by adding or dropping experimental arms—critical features for trials in rapidly developing disease areas such as COVID-19 or cancer therapeutics. However, in these trials, efficacy and safety issues may render certain participant subgroups ineligible to some experimental arms, and methods for stratified randomization do not readily apply to this setting. Methods: We propose extensions for conventional methods of stratified randomization for platform trials whose experimental arms may differ in eligibility criteria. These methods balance on a prespecified set of stratification variables observable prior to arm assignment that remains the same across experimental arms. To do so, we suggest modifying block randomization by including experimental arm eligibility as a stratifying variable, and we suggest modifying the imbalance score calculation in dynamic balancing by performing pairwise comparisons between each eligible experimental arm and standard of care arm participants eligible to that experimental arm. Results: We provide worked examples to illustrate the proposed extensions. In addition, we also provide a formula to quantify the relative efficiency loss of platform trials with varying eligibility compared with trials with non-varying eligibility as measured by the size of the common standard of care arm. Conclusions: This article presents important extensions to conventional methods for stratified randomization in order to facilitate the implementation of platform trials with differing experimental arm eligibility.

scholarly journals Tag-based Weakly-supervised Hashing for Image Retrieval

2018 ◽  
Author(s):  
Ziyu Guan ◽  
Fei Xie ◽  
Wanqing Zhao ◽  
Xiaopeng Wang ◽  
Long Chen ◽  
...  
Keyword(s):  
Image Retrieval ◽  
Optimization Problem ◽  
Training Data ◽  
Fine Tuning ◽  
Second Stage ◽  
Deep Hashing ◽  
Real World Datasets ◽  
Weakly Supervised ◽  
Similarity Relationships ◽  
Two Stages

We are concerned with using user-tagged images to learn proper hashing functions for image retrieval. The benefits are two-fold: (1) we could obtain abundant training data for deep hashing models; (2) tagging data possesses richer semantic information which could help better characterize similarity relationships between images. However, tagging data suffers from noises, vagueness and incompleteness. Different from previous unsupervised or supervised hashing learning, we propose a novel weakly-supervised deep hashing framework which consists of two stages: weakly-supervised pre-training and supervised fine-tuning. The second stage is as usual. In the first stage, rather than performing supervision on tags, the framework introduces a semantic embedding vector (sem-vector) for each image and performs learning of hashing and sem-vectors jointly. By carefully designing the optimization problem, it can well leverage tagging information and image content for hashing learning. The framework is general and does not depend on specific deep hashing methods. Empirical results on real world datasets show that when it is integrated with state-of-art deep hashing methods, the performance increases by 8-10%.

scholarly journals Common substructures and sequence characteristics of sandwich-like proteins from 42 different folds

2020 ◽  
Author(s):  
A.E. Kister
Keyword(s):  
Protein Design ◽  
Structural Classification ◽  
Hydrophobic Residues ◽  
Rational Protein Design ◽  
The Common ◽  
Database Comparison ◽  
Two Stages ◽  
Sequence Characteristics ◽  
Common Sequence

AbstractThis study addresses the following fundamental question: Do sequences of protein domains with sandwich architecture have common sequence characteristics even though they belong to different superfamilies and folds? The analysis was carried out in two stages: determination of substructures in the domains that are common to all sandwich proteins; and detection of common sequence characteristics within the substructures. Analysis of supersecondary structures in domains of proteins revealed two types of four-strand substructures that are common to sandwich proteins. At least one of these common substructures was found in proteins of 42 sandwich-like folds (as per structural classification in the CATH database). Comparison of the sequence fragments corresponding to strands that make up the common substructures revealed specific rules of distribution of hydrophobic residues within these strands. These rules can be conceptualized as grammatical rules of beta protein linguistics. Understanding of the structural and sequence commonalities of sandwich proteins may also be useful for rational protein design.

Sign in / Sign up

Export Citation Format

Share Document