-
Real-Time Object Detection in Occluded Environment with Background Cluttering Effects Using Deep Learning
Authors:
Syed Muhammad Aamir,
Hongbin Ma,
Malak Abid Ali Khan,
Muhammad Aaqib
Abstract:
Detection of small, undetermined moving objects or objects in an occluded environment with a cluttered background is the main problem of computer vision. This greatly affects the detection accuracy of deep learning models. To overcome these problems, we concentrate on deep learning models for real-time detection of cars and tanks in an occluded environment with a cluttered background employing SSD…
▽ More
Detection of small, undetermined moving objects or objects in an occluded environment with a cluttered background is the main problem of computer vision. This greatly affects the detection accuracy of deep learning models. To overcome these problems, we concentrate on deep learning models for real-time detection of cars and tanks in an occluded environment with a cluttered background employing SSD and YOLO algorithms and improved precision of detection and reduce problems faced by these models. The developed method makes the custom dataset and employs a preprocessing technique to clean the noisy dataset. For training the developed model we apply the data augmentation technique to balance and diversify the data. We fine-tuned, trained, and evaluated these models on the established dataset by applying these techniques and highlighting the results we got more accurately than without applying these techniques. The accuracy and frame per second of the SSD-Mobilenet v2 model are higher than YOLO V3 and YOLO V4. Furthermore, by employing various techniques like data enhancement, noise reduction, parameter optimization, and model fusion we improve the effectiveness of detection and recognition. We further added a counting algorithm, and target attributes experimental comparison, and made a graphical user interface system for the developed model with features of object counting, alerts, status, resolution, and frame per second. Subsequently, to justify the importance of the developed method analysis of YOLO V3, V4, and SSD were incorporated. Which resulted in the overall completion of the proposed method.
△ Less
Submitted 1 January, 2024;
originally announced January 2024.
-
Experimental Comparison of SNR and RSSI for LoRa-ESL Based on Machine Clustering and Arithmetic Distribution
Authors:
Malak Abid Ali Khan,
Hongbin Ma,
Syed Muhammad Aamir,
Cekderi Anil Baris
Abstract:
LoRa lacks the sensing capabilities of channel status. Received signal strength indicator (RSSI) decreases due to collision, interference, and near-far effect while for signal-to-noise ratio (SNR), the packets are rejected by decreasing the transmission power (TP) at a higher spreading factor (SF). To overcome these challenges in the case of electric shelf label (ESL) to minimize the dependency on…
▽ More
LoRa lacks the sensing capabilities of channel status. Received signal strength indicator (RSSI) decreases due to collision, interference, and near-far effect while for signal-to-noise ratio (SNR), the packets are rejected by decreasing the transmission power (TP) at a higher spreading factor (SF). To overcome these challenges in the case of electric shelf label (ESL) to minimize the dependency on retransmission and acknowledgment, the end devices (EDs) are allocated around gateways (GWs) based on machine clustering with dynamic SF for SNR while dynamic TP for RSSI. The experimental results determined that the RSSI approach is more dominant than SNR because of determining the exact locality of the ED that diminished the capture effect. Arithmetic distribution of EDs for various GWs in different clusters helps to minify the near-far effect. The resultant received power (RP) at each cluster is higher for most of the connected EDs than the threshold RP.
△ Less
Submitted 13 December, 2022; v1 submitted 26 October, 2022;
originally announced October 2022.
-
Object Recognition in Different Lighting Conditions at Various Angles by Deep Learning Method
Authors:
Imran Khan Mirani,
Chen Tianhua,
Malak Abid Ali Khan,
Syed Muhammad Aamir,
Waseef Menhaj
Abstract:
Existing computer vision and object detection methods strongly rely on neural networks and deep learning. This active research area is used for applications such as autonomous driving, aerial photography, protection, and monitoring. Futuristic object detection methods rely on rectangular, boundary boxes drawn over an object to accurately locate its location. The modern object recognition algorithm…
▽ More
Existing computer vision and object detection methods strongly rely on neural networks and deep learning. This active research area is used for applications such as autonomous driving, aerial photography, protection, and monitoring. Futuristic object detection methods rely on rectangular, boundary boxes drawn over an object to accurately locate its location. The modern object recognition algorithms, however, are vulnerable to multiple factors, such as illumination, occlusion, viewing angle, or camera rotation as well as cost. Therefore, deep learning-based object recognition will significantly increase the recognition speed and compatible external interference. In this study, we use convolutional neural networks (CNN) to recognize items, the neural networks have the advantages of end-to-end, sparse relation, and sharing weights. This article aims to classify the name of the various object based on the position of an object's detected box. Instead, under different distances, we can get recognition results with different confidence. Through this study, we find that this model's accuracy through recognition is mainly influenced by the proportion of objects and the number of samples. When we have a small proportion of an object on camera, then we get higher recognition accuracy; if we have a much small number of samples, we can get greater accuracy in recognition. The epidemic has a great impact on the world economy where designing a cheaper object recognition system is the need of time.
△ Less
Submitted 18 October, 2022;
originally announced October 2022.
-
Structural, Electronic, and Magnetic Properties of HiPIMS Grown Co-N Thin Films
Authors:
Seema,
Akhil Tayal,
S. M. Amir,
Sabine Pütter,
S. Mattauch,
Mukul Gupta
Abstract:
We studied the growth behavior, structural, electronic, and magnetic properties of cobalt nitride (Co-N) thin films deposited using direct current (dc) and high power impulse magnetron sputtering (HiPIMS) processes. The N$_2$ partial gas flow (\pn) was varied in close intervals to achieve the optimum conditions for the growth of tetra cobalt nitride (\tcn) phase. We found that Co-N films grown usi…
▽ More
We studied the growth behavior, structural, electronic, and magnetic properties of cobalt nitride (Co-N) thin films deposited using direct current (dc) and high power impulse magnetron sputtering (HiPIMS) processes. The N$_2$ partial gas flow (\pn) was varied in close intervals to achieve the optimum conditions for the growth of tetra cobalt nitride (\tcn) phase. We found that Co-N films grown using HiPIMS process adopt (111) orientation as compared to the growth taking place along the (100) direction in the dcMS process. It was observed that HiPIMS grown Co-N~films were superior in terms of crystallite size and uniform surface morphology. The local structure of films was investigated using x-ray absorption fine structure (XAFS) measurements. We found that the high energy of adatoms in the HiPIMS technique assisted in the greater stabilization of fcc-Co and novel \tcn~phase relative to the dcMS process. Magnetic properties of Co-N thin films were studied using magneto-optical Kerr effect, vibrating sample magnetometry and polarized neutron reflectivity. It was found that though the saturation magnetization remains almost similar in films grown by dcMS or HiPIMS processes, they differ in terms of their magnetic anisotropy. Such variation can be understood in terms of differences in the growth mechanisms in dcMS and HiPIMS processes affecting the local structure of resulting \tcn~phase.
△ Less
Submitted 23 May, 2020;
originally announced May 2020.
-
Influence of interface and microstructure on magnetization of epitaxial Fe4N thin film
Authors:
Nidhi Pandey,
S. Putter,
S. M. Amir,
V. R. Reddy,
D. M. Phase,
J. Stahn,
Ajay Gupta,
Mukul Gupta
Abstract:
Epitaxial Fe4N thin films grown on lattice-matched LaAlO3 (LAO) substrate using sputtering and molecular beam epitaxy techniques have been studied in this work. Within the sputtering process, films were grown with conventional direct current magnetron sputtering (dcMS) and for the first time, using a high power impulse magnetron sputtering (HiPIMS) process. Surface morphology and depth profile rev…
▽ More
Epitaxial Fe4N thin films grown on lattice-matched LaAlO3 (LAO) substrate using sputtering and molecular beam epitaxy techniques have been studied in this work. Within the sputtering process, films were grown with conventional direct current magnetron sputtering (dcMS) and for the first time, using a high power impulse magnetron sputtering (HiPIMS) process. Surface morphology and depth profile reveal that HiPIMS deposited film has the lowest roughness, the highest packing density and the sharpest interface. La from the LAO substrate and Fe from the film interdiffuse and forms an undesired interface spreading to an extent of about 10-20 nm. In the HiPIMS process, layer by layer type growth leads to a globular microstructure which restricts the extent of the interdiffused interface. Such substrate-film interactions and microstructure play a vital role in affecting the electronic hybridization and magnetic properties of Fe4N films. The magnetic moment (Ms) was compared using bulk, element-specific and magnetic depth profiling techniques. We found that Ms was the highest when the thickness of the interdiffused layer was lowest and only be achieved in the HiPIMS grown samples. Presence of small moment at the N site was also evidenced by element-specific x-ray circular dichroism measurement in HiPIMS grown sample. A large variation in the Ms values of Fe4N found in the experimental works carried out so far could be due to such interdiffused layer which is generally not expected to form in otherwise stable oxide substrate. In addition, a consequence of substrate-film interdiffusion and microstructure results in different kinds of different kind of magnetic anisotropies in films grown using different techniques.
△ Less
Submitted 4 June, 2019;
originally announced June 2019.
-
Surfactant enhanced antiferromagnetic coupling in magnetron sputtered Cu/Co multilayers: A neutron reflectivity study
Authors:
S. M. Amir,
Mukul Gupta,
Ajay Gupta,
J. Stahn
Abstract:
In this work we studied Cu/Co multilayers prepared using dc-magnetron sputtering technique with Ag surfactant. It was found that Ag balances the difference in the surface free energy of Cu and Co and this results in removing the asymmetry in the interface roughness of Cu-on-Co and Co-on-Cu interfaces. As the interfaces become symmetric, we observe a significant enhancement in antiferromagnetic cou…
▽ More
In this work we studied Cu/Co multilayers prepared using dc-magnetron sputtering technique with Ag surfactant. It was found that Ag balances the difference in the surface free energy of Cu and Co and this results in removing the asymmetry in the interface roughness of Cu-on-Co and Co-on-Cu interfaces. As the interfaces become symmetric, we observe a significant enhancement in antiferromagnetic coupling and magneto resistance. Further, a correlation of spin-dependent scattering with the interface roughness is brought by comparing Cu/Co multilayer prepared using different deposition methods. It was found that as interface roughness increases spin-dependent scattering decreases.
△ Less
Submitted 24 September, 2012;
originally announced September 2012.
-
Surfactant induced smooth and symmetric interfaces in Cu/Co multilayers
Authors:
S. M. Amir,
Mukul Gupta,
Ajay Gupta,
J. Stahn,
A. Wildes
Abstract:
In this work we studied Ag surfactant induced growth of Cu/Co multilayers. The Cu/Co multilayers were deposited using Ag surfactant by ion beam sputtering technique. It was found that Ag surfactant balances the asymmetry between the surface free energy of Cu and Co. As a result, the Co-on-Cu and Cu-on-Co interfaces become sharp and symmetric and thereby improve the thermal stability of the multila…
▽ More
In this work we studied Ag surfactant induced growth of Cu/Co multilayers. The Cu/Co multilayers were deposited using Ag surfactant by ion beam sputtering technique. It was found that Ag surfactant balances the asymmetry between the surface free energy of Cu and Co. As a result, the Co-on-Cu and Cu-on-Co interfaces become sharp and symmetric and thereby improve the thermal stability of the multilayer. On the basis of obtained results, a mechanism leading to symmetric and stable interfaces in Cu/Co multilayers is discussed.
△ Less
Submitted 7 February, 2011;
originally announced February 2011.
-
Surfactant mediated growth of Ti/Ni multilayers
Authors:
Mukul Gupta,
S. M. Amir,
Ajay Gupta,
Jochen Stahn
Abstract:
The surfactant mediated growth of Ti/Ni multilayers is studied. They were prepared using ion beam sputtering at different adatom energies. It was found that the interface roughness decreased significantly when the multilayers were sputtered with Ag as surfactant at an ion energy of 0.75 keV. On the other hand, when the ion energy was increased to 1 keV, it resulted in enhanced intermixing at the i…
▽ More
The surfactant mediated growth of Ti/Ni multilayers is studied. They were prepared using ion beam sputtering at different adatom energies. It was found that the interface roughness decreased significantly when the multilayers were sputtered with Ag as surfactant at an ion energy of 0.75 keV. On the other hand, when the ion energy was increased to 1 keV, it resulted in enhanced intermixing at the interfaces and no appreciable effect of Ag surfactant could be observed. On the basis of the obtained results, the influence of adatom energy on the surfactant mediated growth mechanism is discussed.
△ Less
Submitted 3 February, 2011;
originally announced February 2011.
