Hazrat Bilal

University of Science and Technology of China, Automation, Graduate Student

Federal Urdu University of Arts, Science and Technology,Islamabad, Pakistan, Department of Electrical Engineering, Graduate Student

Nanjing University of science and technology, School Of Automation, Graduate Student

Followers

Following

Co-author

Public Views

@PhD Research Scholar
@Robotics and Artificial Intelligence
Supervisors: Prof. Baoqun Yin
Phone: 13390916272
Address: USTC, Hefei, Anhui Province, China

less

VISHAL PATEL

Riccardo CONTINO

University of Salento

Dawei Du

Rajeev Yasarla

Abdulfattah Noorwali

Western University Canada

Gonzalo Montalva

Dawei Du

Joost van de Weijer

Universitat Autònoma de Barcelona

Chengjiang Long

Juan Carlos Padilla Osorio

Interests

Uploads

Papers by Hazrat Bilal

Second-Order Convolutional Network for Crowd Counting

Fourth International Workshop on Pattern Recognition,, 2019

Single image crowd counting remains challenging primarily due to various issues, such as large sc... more Single image crowd counting remains challenging primarily due to various issues, such as large scale variations, perspective and non-uniform crowd distribution. In this paper, we propose a novel architecture referred to Second-Order Convolutional Network (SOCN) to deal with this task from the perspective of improving the feature transformation capability of the network. The proposed SOCN applies a convolutional neural network as the backbone. We introduce three cascaded second-order blocks located behind the backbone to augment the family of transformation operations and increase the nonlinearity of the network, which can extract multi-scale and discriminative features. Furthermore, we design a context attention module (CAM) including dilated convolutions to assign weights to the score map of each second-order block for the purpose that the features which contribute to counting can be highlighted. We conduct various experiments on ShanghaiTeach 1 and UCF_CC_50 2 datasets, and the results demonstrate the effectiveness of our method.

Download

Reduction of Multiplications in Convolutional Neural Networks

by Hazrat Bilal and Munawar Ali

Chinese Control Conference, 2020

Convolution neural network (CNN) widely used for the application of computer vision, such applica... more Convolution neural network (CNN) widely used for the application of computer vision, such applications would be beneficial for CNN if their workload could be reduced. In this paper, we evaluate CNN's arithmetic properties (Matrix Multiplications and Additions) and proposed a Strassen algorithm with the help of Pan's result to minimize their workload of computation. That algorithm needs the least workspace and it has great computational accuracy. Matrix multiplication (MM) is the most fundamental computational operation, and the performance of Matrix multiplication (MM) depends on different factors. Moreover, that is a very flexible and robust structure. Here we are focusing on the number of element-wise multiplications and addition, the effects of this research on modern computations.

Download

Real-Time Lane Detection and Tracking for Advanced Driver Assistance Systems

Chinese Control Conference , 2019

Due to the fast-growing industry of intelligent vehicles the advanced driver assistance system (A... more Due to the fast-growing industry of intelligent vehicles the advanced driver assistance system (ADAS) has engrossed a lot of attention of the scholars. One of the biggest hurdles for new autonomous vehicles is to detect curvy lanes, multiple lanes, and lanes with a lot of discontinuity and noise. The purpose of this paper is to analyze the possibilities of image processing techniques for a computer vision application focusing on the problem of lane detection to enable traffic safety and driving comfort. The proposed algorithm is a combination of two sub-algorithms. The first sub-algorithm called Fuzzy Noise Reduction Filter (FNRF); removes the noise and smoothen the sequences of images received by the camera. While the second sub-algorithm aims to detect lane in normal as well as challenging scenarios by applying the concept of Hough Transform (HT) with a capable region of interest. The novelty of the proposed research study is the tracking of the lanes under inclement weather and challenging lightening conditions with improved computational time. The result achieved through our proposed algorithm is satisfactory in video sequences captured on several road types and under very challenging lighting and weather conditions.

Download

Pruning filters with L1-norm and capped L1-norm for CNN compression

Applied Intelligence, 2020

The blistering progress of convolutional neural networks (CNNs) in numerous applications of the r... more The blistering progress of convolutional neural networks (CNNs) in numerous applications of the real-world usually obstruct by a surge in network volume and computational cost. Recently, researchers concentrate on eliminating these issues by compressing the CNN models, such as pruning filters and weights. In comparison with the technique of pruning weights, the technique of pruning filters doesn't effect in sparse connectivity patterns. In this article, we have proposed a fresh new technique to estimate the significance of filters. More precisely, we combined L1-norm with capped L1-norm to represent the amount of information extracted by the filter and control regularization. In the process of pruning, the insignificant filters remove directly without any loss in the test accuracy, providing much slimmer and compact models with comparable accuracy and this process is iterated a few times. To validate the effectiveness of our algorithm. We experimentally determine the usefulness of our approach with several advanced CNN models on numerous standard data sets. Particularly, data sets CIFAR-10 is used on VGG-16 and prunes 92.7% parameters with float-point-operations (FLOPs) reduction of 75.8% without loss of accuracy and has achieved advancement in state-of-art.

Download

Jerk-bounded trajectory planning for rotary flexible joint manipulator: an experimental approach

Soft Computing SPRINGER, 2023

The fundamental criteria for industrial manipulator applications are vibration free and smooth mo... more The fundamental criteria for industrial manipulator applications are vibration free and smooth motion with minimum time. This paper investigates the trajectory tracking and vibration control of rotary flexible joint manipulator with parametric uncertainties. Firstly, the dynamic modeling via Euler Lagrange equation for a single link flexible joint manipulator is discussed. Secondly, for the execution of smooth motion between two points, bounded and continuous jerk trajectory is developed and implemented. In addition, the prospective strategy uses the concatenation of fifth-order polynomials to provide a smooth trajectory between two-way points. In the planned algorithm, user can independently define the position, velocity, acceleration and jerk values at both initial and final positions. The feature of user-defined parameters gives the versatility to the suggested algorithm for generating trajectories for diverse applications of robotic manipulators. Moreover, the planned scheme is easy to implement and computationally efficient. In the last, the performance of the presented scheme is examined by comparison with cubic splines and a linear segment with parabolic blends (LSPB) techniques. Generated trajectories were evaluated successfully by carrying multiple experiments on QUANSER's flexible joint manipulator.

Download

A practical study of active disturbance rejection control for rotary flexible joint robot manipulator

Soft Computing; SPRINGER, 2023

This research presents a practical study of active disturbance rejection control law (ADRC) for t... more This research presents a practical study of active disturbance rejection control law (ADRC) for the control of robotic manipulators in the presence of uncertainties. The control objective of the proposed control law is to track the trajectory accurately and rejects the disturbance caused by robotic manipulators. First, Euler Lagrange's equations are used to model the dynamic behavior of a rotary flexible joint manipulator system (RFJMS). Secondly, the ADRC is designed for the rejection of lump disturbances (modeling uncertainties, nonlinearities, and external disturbances) of the manipulator; to estimate the lump disturbances a fifth-order extended state observer is constructed. Furthermore, a state error feedback control law with disturbance compensation is designed, which needs only a few control parameters to be adjusted. Furthermore, we demonstrate the robustness and effectiveness of the proposed control law by comparing the experimental results with those of LQR and PID. Experimental results show that high precision position control as well as vibration suppression can be achieved with the proposed controller, which is superior to LQR and PID controllers. In spite of uncertainties and nonlinearity of the flexible joint, QUANSER's RFJMS exhibits excellent tracking behavior and disturbance rejection.

Download

Experimental Validation of Fuzzy PID Control of Flexible Joint System in Presence of Uncertainties

The 36th IEEE Chinese Control Conference (CCC), 2017

Abstract:In this paper, a robust fuzzy-tuned PID controller is proposed for the trajectory tracki... more Abstract:In this paper, a robust fuzzy-tuned PID controller is proposed for the trajectory tracking and vibration control of a flexiblejoint
manipulator system (FJMS) with parametric uncertainties. Flexible joint manipulators are widely used in industry due to their flexibility,
light in weight and fast response, low cost, and low energy consumption. The dynamic modeling via Euler-Lagrange equation for asingle
link flexible joint manipulator system (FJMS) is presented first. By analyzing the characteristics of the high order linear system, the
Fuzzy-Tuned PID control strategy is utilized. The parameters of classical PID controller are updated by the intelligent method of fuzzy
logic. With the adaptivity to the parametric uncertainties from the system stiffness and moment of inertia, the contrived approach provides
more efficient results compared to existing techniques in terms of vibration suppressions, time response, and overshoot. Finally, the
effectiveness of the controller is corroborated through multiple experiments on the QUANSER’s flexible joint manipulator system
(FJMS).
Key Words:Fuzzy Tuned PID Control, Vibration Suppressions, Trajectory Tracking, Flexible Joint Manipulator.

Download

Second-Order Convolutional Network for Crowd Counting

Fourth International Workshop on Pattern Recognition,, 2019

Download

Reduction of Multiplications in Convolutional Neural Networks

by Hazrat Bilal and Munawar Ali

Chinese Control Conference, 2020

Download

Real-Time Lane Detection and Tracking for Advanced Driver Assistance Systems

Chinese Control Conference , 2019

Download

Pruning filters with L1-norm and capped L1-norm for CNN compression