Pedestrian Detection and Tracking Using Deformable Part Models and Kalman Filtering

Abstract： Pedestrian detection techniques are important and challenging especially for complex real world scenes. They can be used for ensuring pedestrian safety, ADASs （advance driver assistance systems） and safety surveillance systems. In this paper, we propose a novel approach for multi-person tracking-by-detection using deformable part models in Kalman filtering framework. The Kalman filter is used to keep track of each person and a unique label is assigned to each tracked individual. Based on this approach, people can enter and leave the scene randomly. We test and demonstrate our results on Caltech Pedestrian benchmark, which is two orders of magnitude larger than any other existing datasets and consists of pedestrians varying widely in appearance, pose and scale. Complex situations such as people occluded by each other are handled gracefully and individual persons can be tracked correctly after a group of people split. Experiments confirm the real-time performance and robustness of our system, working in complex scenes. Our tracking model gives a tracking accuracy of 72.8% and a tracking precision of 82.3%. We can further reduce false positives by 2.8%, using Kalman filtering.     

An efficient implicit regularized Lagrangian twin support vector regression

Twin support vector regression (TSVR) and Lagrangian TSVR (LTSVR) satisfy only empirical risk minimization principle. Moreover, the matrices in their formulations are always positive semi-definite. To overcome these problems, we propose an efficient implicit Lagrangian formulation for the dual regularized twin support vector regression, called IRLTSVR for short. By introducing a regularization term to each objective function, the optimization problems in our IRLTSVR are positive definite and implement the structural risk minimization principle. Moreover, the 1-norm of the vector of slack variable is replaced with 2-norm to make the objective functions strongly convex. Our IRLTSVR solves two systems of linear equations instead of solving two quadratic programming problems (QPPs) in TSVR and one large QPP in SVR, which makes the learning speed of IRLTSVR faster than TSVR and SVR. Particularly, we compare three implementations of IRLTSVR with existing approaches. Computational results on several synthetic and real-world benchmark datasets clearly indicate the effectiveness and applicability of the IRLTSVR in comparison to SVR, TSVR and LTSVR.     

Seismic response of precast reinforced concrete wall subjected to cyclic in-plane and constant out-of-plane loading

This paper provides insight into the seismic behavior of a full-scale precast reinforced concrete wall under in-plane cyclic loading combined with out-of-plane loading replicated by sand backfill to simulate the actual condition of basement walls. The tested wall exhibited flexural cracks, owing to the high aspect ratio and considerable out-of-plane movement due to lateral pressure from the backfill. The wall performed satisfactorily by exhibiting competent seismic parameters and deformation characteristics governed by its ductile response in the nonlinear phase during the test with smaller residual drift. Numerical analysis was conducted to validate experimental findings, which complied with each other. The numerical model was used to conduct parametric studies to study the effect of backfill density and aspect ratio on seismic response of the proposed precast wall system. The in-plane capacity of walls reduced, while deformation characteristics were unaffected by the increase in backfill density. An increase in aspect ratio leads to a reduction in in-plane capacity and an increase in drift. Curves between the ratio of in-plane yield capacity and design shear load of walls are proposed for the backfill density, which may be adopted to determine the in-plane yield capacity of the basement walls based on their design shear.     

Influence of soil density on gas permeability and water retention in soils amended with in-house produced biochar

Biochar has been used as an environment-friendly enhancer to improve the hydraulic properties(e.g.suction and water retention) of soil.However,variations in densities alter the properties of the soil-biochar mix.Such density variations are observed in agriculture(loosely compacted) and engineering(densely compacted) applications.The influence of biochar amendment on gas permeability of soil has been barely investigated,especially for soil with diffe rent densities.The maj or obj ective of this study is to investigate the water retention capacity,and gas permeability of biochar-amended soil(BAS) with different biochar contents under varying degree of compaction(DOC) conditions.In-house produced novel biochar was mixed with the soil at different amendment rates(i.e.biochar contents of 0%,5% and 10%).All BAS samples were compacted at three DOCs(65%,80% and 95%) in polyvinyl chloride(PVC)tubes.Each soil column was subjected to drying-wetting cycles,during which soil suction,water content,and gas permeability were measured.A simplified theoretical framework for estimating the void ratio of BAS was proposed.The experimental results reveal that the addition of biochar significantly decreased gas permeability k_g as compared with that of bare soil(BS).However,the addition of 5%biochar is found to be optimum in decreasing kg with an increase of DOC(i.e.k_(g,65%) k_(g,80%) k_(g,95%)) at a relatively low suction range(200 kPa) because both biochar and compaction treatment reduce the connected pores.     

Neural network based non-invasive method to detect anemia from images of eye conjunctiva

Detection of anemia can be done by examining the hemoglobin concentration level in the blood using complete blood count, which is an invasive, time-consuming, and costly technique. Preliminary methods for detecting anemia include examining the color of the palpebral conjunctiva, which is a non-invasive method, but color perception may vary from person to person. This study aims to develop a computerized non-invasive technique for anemia detection. We propose a novel machine learning model using the artificial neural network to detect anemic patients from the images of eye conjunctiva. Since limited and small dataset has been used in the earlier approaches, this may cause over fitting of the model. We have improved the number of available training images using image augmentation techniques. To standardize a non-invasive method, we have used computer vision algorithms for preprocessing and feature extraction. This article derives the backpropagation rules mathematically for adjusting the weights for the proposed neural network model. After hyper parameter tuning and using the mathematically derived backpropagation rules, the model was able to achieve the best accuracy of 97.00% with sensitivity 99.21% and specificity 95.42% on the created dataset.     

Synthesis of CuTi-LDH supported on g-C3N4 for electrochemical and photoelectrochemical oxygen evolution reactions

A nanocomposite CuTi layered double hydroxide (LDH) supported on g-C₃N₄ (15 wt% of g-C₃N₄) is facilely synthesized by hydrothermal method. There are electrostatic interactions between positive layers of CuTi-LDH and negatively charged inner g-C₃N₄ sheets. The nanocomposite and its precursors are characterized through various analytical techniques, which affirmed the presence of both g-C₃N₄ and CuTi-LDH characteristic features. The pore-enriched hybrid geometry of CuTi-LDH@g-C₃N₄ with high specific surface area (146 m²/g), and suitable band gap of 2.46 eV enables the nanocomposite to act as both an electrocatalyst and photoelectrocatalyst for oxygen evolution reaction (OER). Both the electrochemical and photoelectrochemical studies are done using 1 M KOH (pH = 13.6) with applied potential of ?0.2 V to 1.5 V vs. Ag/AgCl. The onset potential of CuTi-LDH@g-C₃N₄ for OER appears at η = 0.36 V in dark and η = 0.32 V under visible light illumination of 30 min. Also, Mott-Schottky analysis shows n-type semiconductor behaviour for CuTi-LDH@g-C₃N₄ and its precursors. The photoelectrochemical water oxidation proceeds by charge transfer across a Type II heterojunction formed between the CuTi-LDH and g-C₃N₄ materials.     

Performance Analysis of Ferroelectric GAA MOSFET with Metal Grain Work Function Variability

Silicon - This work represents a unique GAA MOSFET with metal work-function variations (WFVs) and ferroelectric material as dielectric. A random distribution of metal grain (TiN) with grain size...     

Investigation on DC/RF Performance of LG = 19 nm Heterogeneous Integrated Ga0.15In0.85As/InAs/Ga0.15In0.85As Composite Channel InP HEMT on Silicon Substrate for Future Beyond 5G and Quantum Computing Applications

Silicon - The RF/DC performances of LG = 19 nm heterogeneous integrated Ga0.15In0.85As/InAs/Ga0.15In0.85As composite channel based InP HEMT (high electron mobility transistor)...     

An Adaptive Non Reference Anchor Array Framework for Audio Retrieval in Teleconferencing Environment

In this paper, an adaptive framework for audio retrieval in live teleconferencing environments with multiple participants is proposed. The framework uses a non reference anchor array (NRA) to capture the interfering speech sources, in addition to the primary array that captures the speech source of interest (SOI). A linearly constrained-minimum variance (LC-MV) beamformer is used herein such that the signal coming from the look direction is preserved while interferences coming from the non look direction are nulled. Additionally, the reverberant component of the speech acquired by this framework is removed by a novel method that uses the linear prediction (LP) residual cepstrum. This method does not require the computation of the acoustic impulse response (AIR) of the teleconferencing room and hence is computationally efficient. The NRA framework is therefore able to remove correlated noise coming from the direction of the SOI and also dereverberating the noise free signal. The performance of the proposed framework is evaluated by conducting experiments on clean speech acquisition from distant microphone arrays. Experiments on distant speech recognition are also conducted using the TIMIT and MONC databases. Experimental results obtained from the proposed framework indicate a reasonable improvement over correlation, subspace and standard minimum variance beamforming methods. The application of the framework in audio retrieval in a live teleconferencing environment with multiple participants is also discussed.