|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Detailed investigation of optimized alkali catalyzed transesterification of Jatropha oil for biodiesel production显示文摘The non-edible oils are believed to be one of the major feedstock for the production of biodiesel in future.In the present study,we investigated the production of Jatropha oil methyl esters(JOMEs) via alkali-catalyzed transesterification route.The biophysical characteristics of Jatropha oil were found within the optimal range in accordance with ASTM standards as a substitute diesel fuel.The chemical composition and production yield of as-synthesized biodiesel were confirmed by various analytical techniques such as FT-IR,1H NMR,13 C NMR and gas chromatography coupled with mass spectrometry.A high percentage conversion,~96.09%,of fatty acids into esters was achieved under optimized transesterification conditions with 6 :1 oil to methanol ratio and 0.9 wt% Na OH for 50 min at ~60°C.Moreover,twelve fatty acids methyl esters(FAME) were quantified in the GC/MS analysis and it was interesting to note that the mass fragmentation pattern of saturated,monounsaturated and diunsaturated FAME was comparable with the literature reported values. | Waqas Ahmed Muhammad Faizan Nazar Syed Danish Ali Usman Ali Rana Salah Ud-Din Khan | 2015 | Journal of Energy Chemistry2015，24，3： | 5 |
| 2 | Loan growth and bank solvency:evidence from the Pakistani banking sector显示文摘Background:The dramatic loan growth and changes in the Pakistani banking system in mid-2000s have led to significant research attention on borrowers and lenders.This expansion and diversification in financial sector was driven by structural reforms,political stability and significant economic growth.Against this background,this study investigates the loan growth and risk-taking behavior of the banks during the expansionary periods of lending.Method:This study used dynamic two-step system generalized method of moment’s estimation technique,based on data taken from 32 banks in Pakistan over 2006-2014.Result:Loan growth has a significant effect on bank-specific and macroeconomicspecific variables.Loan growth in the previous year raises non-performing loans and decreases the solvency of banks with a time lag of many years.The driving force behind this phenomenon is weak prudential regulation among competitors,the asymmetric information of the borrowers,and,most importantly,that banks underestimate the risk of lending during credit booms.Conclusion:More regulatory measures are required to ensure a strong financial system when the volume of non-performing loan grows significantly.An increase in the capital requirement policy for rapidly growing banks is also needed because the problem of abnormal loan growth cannot be detected at the current time.At the same time,strong supervision is necessary to avoid the adverse consequences of borrower selection. | Muhammad Kashif Syed Faizan Iftikhar Khurram Iftikhar | 2016 | Financial Innovation2016，2，1： | 2 |
| 3 | Exploration of B-site alloying in partially reducing Pb toxicity and regulating thermodynamic stability and electronic properties of halide perovskites显示文摘Alloying strategies provide a high degree of freedom for reducing lead toxicity,improving thermodynamic stability, tuning the optoelectronic properties of ABX3 halide perovskites by varying the alloying element species and their contents.Given the key role of B-site cations in contributing band edge states and modulating structure factors in halide perovskites,the partial replacement of Pb2+with different B-site metal ions has been proposed.Although several experimental attempts have been made to date,the effect of B-site alloying on the stability and electronic properties of halide perovskites has not been fully explored.Herein,we take cubic CsPbBr3 perovskite as the prototype material and systematically explore the effects of B-site alloying on Pb-containing perovskites.According to the presence or absence of the corresponding perovskite phase,the ten alloying elements investigated are classified into three types(i.e.,Type Ⅰ:Sn Ge,Ca,Sr;Type Ⅱ:Cd,Mg,Mn;Type Ⅲ:Ba,Zn,Cu).Based on the first-principles calculations,we obtain the following conclusions.First,these B-site alloys will exist as disordered solid solutions rather than ordered structures at room temperature throughout the composition space.Second,the alloying of Sn and Ge enhances the thermodynamic stability of the cubic perovskite host,whereas the alloying of the other elements has no remarkable effect on the thermodynamic stability of the cubic perovskite host.Third,the underlying physical mechanism for bandgap tuning can be attributed to the atomic orbital energy mismatch or quantum confinement effect.Fourth,the alloying of different elements demonstrates the diversity in the regulation of crystal structure and electronic properties,indicating potential applications in photovoltaic s and self-trapped exciton-based light-emitting applications.Our work provides theoretical guidance for using alloying strategies to reduce lead toxicity,enhance stability,and optimize the electronic properties of halide perovskites to meet the needs of optoelectronic applications. | Xinjiang Wang Muhammad Faizan Kun Zhou Hongshuai Zou Qiaoling Xu Yuhao Fu Lijun Zhang | 2023 | Science China(Physics,Mechanics & Astronomy)2023，66，3： | 1 |
| 4 | Lightweight authentication protocol for e-health clouds in IoT-based applications through 5G technology显示文摘Modem information technology has been utilized progressively to store and distribute a large amount of healthcare data to reduce costs and improve medical facilities.In this context,the emergence of e-Health clouds offers novel opportunities,like easy and remote accessibility of medical data.However,this achievement produces plenty of new risks and challenges like how to provide integrity,security,and confidentiality to the highly susceptible e-Health data.Among these challenges,authentication is a major issue that ensures that the susceptible medical data in clouds is not available to illegal participants.The smart card,password and biometrics are three factors of authentication which fulfill the requirement of giving high security.Numerous three-factor ECC-based authentication protocols on e-Health clouds have been presented so far.However,most of the protocols have serious security flaws and produce high computation and communication overheads.Therefore,we introduce a novel protocol for the e-Health cloud,which thwarts some major attacks,such as user anonymity,offline password guessing,impersonation,and stolen smart card attacks.Moreover,we evaluate our protocol through formal security analysis using the Random Oracle Model(ROM).The analysis shows that our proposed protocol is more efficient than many existing protocols in terms of computation and communication costs.Thus,our proposed protocol is proved to be more efficient,robust and secure. | Minahil Muhammad Faizan Ayub Khalid Mahmood Saru Kumari Arun Kumar Sangaiah | 2021 | Digital Communications and Networks2021，7，2： | 1 |
| 5 | An enhanced scheme for mutual authentication for healthcare services显示文摘With the advent of state-of-art technologies,the Telecare Medicine Information System(TMIS)now offers fast and convenient healthcare services to patients at their doorsteps.However,this architecture engenders new risks and challenges to patients'and the server's confidentiality,integrity and security.In order to avoid any resource abuse and malicious attack,employing an authentication scheme is widely considered as the most effective approach for the TMIS to verify the legitimacy of patients and the server.Therefore,several authentication protocols have been proposed to this end.Very recently,Chaudhry et al.identified that there are vulnerabilities of impersonation attacks in Islam et al.'s scheme.Therefore,they introduced an improved protocol to mitigate those security flaws.Later,Qiu et al.proved that these schemes are vulnerable to the man-in-the-middle,impersonation and offline password guessing attacks.Thus,they introduced an improved scheme based on the fuzzy verifier techniques,which overcome all the security flaws of Chaudhry et al.'s scheme.However,there are still some security flaws in Qiu et al.'s protocol.In this article,we prove that Qiu et al.'s protocol has an incorrect notion of perfect user anonymity and is vulnerable to user impersonation attacks.Therefore,we introduce an improved protocol for authentication,which reduces all the security flaws of Qiu et al.'s protocol.We also make a comparison of our protocol with related protocols,which shows that our introduced protocol is more secure and efficient than previous protocols. | Salman Shamshad Muhammad Faizan Ayub Khalid Mahmood Saru Kumari Shehzad Ashraf Chaudhry Chien-Ming Chen | 2022 | Digital Communications and Networks2022，8，2： | 1 |
| 6 | Development ofa polygonum minus cell suspension culture system and analysis of secondary metabolites enhanced by elicitation显示文摘 | Muhammad Faizan A Shukor Ismanizan Ismail | 2013 | Acta Physiologiae Plantarum2013，35，5： | 1 |
| 7 | Rare earth metal based DES assisted the VPO synthesis for n-butane selective oxidation toward maleic anhydride显示文摘Deep eutectic solvents(DESs) are now considered a new class of ionic liquid analogs that have been generously used in various fields.Herein, vanadium phosphorus oxide(VPO) catalysts are synthesized in combination with a deep eutectic solvent containing rare earth metal(rE-DES), and their catalytic performance in n-butane selective oxidation to produce maleic anhydride(MA) is evaluated. The rE-DES is produced from the interaction of choline chloride(ChCl) and rare earth metal salts(Cerium, Europium, Lanthanum, and Samarium metal salt)(ChCl:rE = 1:0.5–1:3) under mild conditions. It was found that DESs served as structural modifiers and electronic promoters during VPO synthesis. It regulated the chemical state of the catalyst surface, such as the vanadium valence state, acid-base properties, and ratios of V^(4+)/V^(5+),Lat–O/Sur–O and P/V. Various characterization techniques, such as FT-IR, DSC, XRD, SEM, EDS, TEM, Raman, TGA, NH3-TPD, and XPS,were used to examine its physical and chemical characteristics. These characteristics were correlated with the catalytic performance. The VPO catalyst modified by rE-DES showed a significant enhancement of n-butane conversion and MA selectivity while suppressing the selectivity of CO and CO_(2)as well as the CO/CO_(2)ratio compared to the unpromoted VPO catalyst. Especially for Ce-DES-VPO, it increased the n-butane conversion and MA mass yield up to approximately 11% and 10%, respectively. In addition, we evaluated the catalytic performance under different activation atmospheres. | Muhammad Faizan Yingwei Li Xingsheng Wang Piao Song Ruirui Zhang Ruixia Liu | 2023 | Green Energy & Environment2023，8，6： | 0 |
| 8 | 850 nm centered wavelength-swept laser based on a wavelength selection galvo filter显示文摘A wavelength-swept laser is constructed using a free space external cavity configuration coupled with a fiberbased ring cavity at the 850 nm region. The external cavity filter employs a galvo-mirror scanner with a diffraction grating for wavelength selection. The filter is connected to a ring cavity through an optical circulator.The ring cavity contains a broadband semiconductor optical amplifier with a high optical output. The performance of this laser is demonstrated with broad bandwidths and narrow linewidths. The 3 dB linewidth and the bandwidth of this source are 0.05 nm(~20 GHz) and 48 nm, respectively. The maximum output power is 26 m W at 160 m A current. | Muhammad Faizan Shirazi Mansik Jeon Jeehyun Kim | 2016 | Chinese Optics Letters2016，14，1： | 0 |
| 9 | Development of 19-plex Y STR system and polymorphism studies in Pakistani population显示文摘For the development of 19-plex Y STR system and polymorphism studies in local ethnic populations sixteen markers of non-recombining regions (NRY) of Y chromosome, which show high power of discrimination among individuals, were selected in this study. Blood samples (600) were collected from the males of three most common castes of Pakistani population (Arain, Awan and Rajput) with different parent lineages. Three markers (DYS385a/b, DYS389I/II and YCAIIa/b) among 16 Y STRs are double-targeted regions of the Y chromosome and thus provide two polymorphic peaks for each respective primer set. These 16 Y-STRs were developed into Megaplex system for simultaneous amplification of all markers within the population. The overall power of discrimination observed in focused populations was 60.5%, 66.5% and 55% in Rajput, Awan and Arain casts respectively. This discrimination power will be helpful in human identification for forensic casework studies including sexual assaults and paternity testing. | Faraz Malik Mahmood A. Kayani M. Ansar Obaid Ullah Muhammad Shafeeq Shahid Chohan Yassir Abbas Saqib Shazad Ali Raza Rahat Rehman Faizan Raiz Qurat-ul-ain Muhammad Hassan Siddiqi Allah Rakha Zia ur Rehman Zahoor Ahmed | 2008 | Journal of Pharmaceutical Analysis2008，20，4： | 0 |
| 10 | Smart MobiNet:A Deep Learning Approach for Accurate Skin Cancer Diagnosis显示文摘The early detection of skin cancer,particularly melanoma,presents a substantial risk to human health.This study aims to examine the necessity of implementing efficient early detection systems through the utilization of deep learning techniques.Nevertheless,the existing methods exhibit certain constraints in terms of accessibility,diagnostic precision,data availability,and scalability.To address these obstacles,we put out a lightweight model known as Smart MobiNet,which is derived from MobileNet and incorporates additional distinctive attributes.The model utilizes a multi-scale feature extraction methodology by using various convolutional layers.The ISIC 2019 dataset,sourced from the International Skin Imaging Collaboration,is employed in this study.Traditional data augmentation approaches are implemented to address the issue of model overfitting.In this study,we conduct experiments to evaluate and compare the performance of three different models,namely CNN,MobileNet,and Smart MobiNet,in the task of skin cancer detection.The findings of our study indicate that the proposed model outperforms other architectures,achieving an accuracy of 0.89.Furthermore,the model exhibits balanced precision,sensitivity,and F1 scores,all measuring at 0.90.This model serves as a vital instrument that assists clinicians efficiently and precisely detecting skin cancer. | Muhammad Suleman Faizan Ullah Ghadah Aldehim Dilawar Shah Mohammad Abrar Asma Irshad Sarra Ayouni | 2023 | Computers, Materials & Continua2023，77，12： | 0 |
| 11 | Progress of vanadium phosphorous oxide catalyst for n-butane selective oxidation显示文摘The utilization of lighter alkanes into useful chemical products is essential for modern chemistry and reducing the CO_(2)emission.Particularly,n-butane has gained special attention across the globe due to the abundant production of maleic anhydride(MA).Vanadium phosphorous oxide(VPO)is the most effective catalyst for selective oxidation of n-butane to MA so far.Interestingly,the VPO complex exists in more or less fifteen different structures,each one having distinct phase composition and exclusive surface morphology and physiochemical properties such as valence state,lattice oxygen,acidity etc.,which relies on precursor preparation method and the activation conditions of catalysts.The catalytic performance of VPO catalyst is improved by adding different promoters or co-catalyst such as various metals dopants,or either introducing template or structural-directing agents.Meanwhile,new preparation strategies such as electrospinning,ball milling,hydrothermal,barothermal,ultrasound,microwave irradiation,calcination,sol-gel method and solvothermal synthesis are also employed for introducing improvement in catalytic performance.Research in above-mentioned different aspects will be ascribed in current review in addition to summarizing overall catalysis activity and final yield.To analyze the performance of the catalytic precursor,the reaction mechanism and reaction kinetics both are discussed in this review to help clarify the key issues such as strong exothermic reaction,phosphorus supplement,water supplement,deactivation,and air/n-butane pretreatment etc.related to the various industrial applications of VPO. | Muhammad Faizan Yingwei Li Ruirui Zhang Xingsheng Wang Piao Song Ruixia Liu | 2022 | Chinese Journal of Chemical Engineering2022，35，3： | 0 |
| 12 | An Innovative Approach Utilizing Binary-View Transformer for Speech Recognition Task显示文摘The deep learning advancements have greatly improved the performance of speech recognition systems,and most recent systems are based on the Recurrent Neural Network(RNN).Overall,the RNN works fine with the small sequence data,but suffers from the gradient vanishing problem in case of large sequence.The transformer networks have neutralized this issue and have shown state-of-the-art results on sequential or speech-related data.Generally,in speech recognition,the input audio is converted into an image using Mel-spectrogram to illustrate frequencies and intensities.The image is classified by the machine learning mechanism to generate a classification transcript.However,the audio frequency in the image has low resolution and causing inaccurate predictions.This paper presents a novel end-to-end binary view transformer-based architecture for speech recognition to cope with the frequency resolution problem.Firstly,the input audio signal is transformed into a 2D image using Mel-spectrogram.Secondly,the modified universal transformers utilize the multi-head attention to derive contextual information and derive different speech-related features.Moreover,a feedforward neural network is also deployed for classification.The proposed system has generated robust results on Google’s speech command dataset with an accuracy of 95.16%and with minimal loss.The binary-view transformer eradicates the eventuality of the over-fitting problem by deploying a multiview mechanism to diversify the input data,and multi-head attention captures multiple contexts from the data’s feature map. | Muhammad Babar Kamal Arfat Ahmad Khan Faizan Ahmed Khan Malik Muhammad Ali Shahid Chitapong Wechtaisong Muhammad Daud Kamal Muhammad Junaid Ali Peerapong Uthansakul | 2022 | Computers, Materials & Continua2022，，9： | 0 |
| 13 | Crop Yield Prediction Using Machine Learning Approaches on a Wide Spectrum显示文摘The exponential growth of population in developing countries likeIndia should focus on innovative technologies in the Agricultural processto meet the future crisis. One of the vital tasks is the crop yield predictionat its early stage;because it forms one of the most challenging tasks inprecision agriculture as it demands a deep understanding of the growth patternwith the highly nonlinear parameters. Environmental parameters like rainfall,temperature, humidity, and management practices like fertilizers, pesticides,irrigation are very dynamic in approach and vary from field to field. In theproposed work, the data were collected from paddy fields of 28 districts in widespectrum of Tamilnadu over a period of 18 years. The Statistical model MultiLinear Regression was used as a benchmark for crop yield prediction, whichyielded an accuracy of 82% owing to its wide ranging input data. Therefore,machine learning models are developed to obtain improved accuracy, namelyBack Propagation Neural Network (BPNN), Support Vector Machine, andGeneral Regression Neural Networks with the given data set. Results showthat GRNN has greater accuracy of 97% (R2 = 0.97) with a normalizedmean square error (NMSE) of 0.03. Hence GRNN can be used for crop yieldprediction in diversified geographical fields. | SVinson Joshua ASelwin Mich Priyadharson Raju Kannadasan Arfat Ahmad Khan Worawat Lawanont Faizan Ahmed Khan Ateeq Ur Rehman Muhammad Junaid Ali | 2022 | Computers, Materials & Continua2022，，9： | 0 |
| 14 | Brain Tumor Segmentation using Multi-View Attention based Ensemble Network显示文摘Astrocytoma IV or glioblastoma is one of the fatal and dangerous types of brain tumors.Early detection of brain tumor increases the survival rate and helps in reducing the fatality rate.Various imaging modalities have been used for diagnosing by expert radiologists,and Medical Resonance Image(MRI)is considered a better option for detecting brain tumors as MRI is a non-invasive technique and provides better visualization of the brain region.One of the challenging issues is to identify the tumorous region from the MRI scans correctly.Manual segmentation is performed by medical experts,which is a time-consuming task and got chances of errors.To overcome this issue,automatic segmentation is performed for quick and accurate results.The proposed approach is to capture inter-slice information and reduce the outliers.Deep learning-based brain tumor segmentation techniques proved best among available segmentation techniques.However,deep learning may miss some preliminary info while using MRI images during segmentation.As MRI volumes are volumetric,3D U-Net-based models are used but complex.Combinations of multiple 2D U-Net predictions in axial,sagittal,and coronal views help to capture inter-slice information.This approach may reduce the system complexity.Moreover,the Conditional Random Fields(CRF)reduce the predictions’false positives and improve the segmentation results.This model is applied to Brain Tumor Segmentation(BraTS)2019 dataset,and cross-validation is performed to check the accuracy of results.The proposed approach achieves Dice Similarity Score(DSC)of 0.77 on Enhancing Tumor(ET),0.90 on Whole Tumor(WT),and 0.84 on Tumor Core(TC)with reduced Hausdorff Distance(HD)of 3.05 on ET,5.12 on WT and 3.89 on TC. | Noreen Mushtaq Arfat Ahmad Khan Faizan Ahmed Khan Muhammad Junaid Ali Malik Muhammad Ali Shahid Chitapong Wechtaisong Peerapong Uthansakul | 2022 | Computers, Materials & Continua2022，，9： | 0 |
| 15 | Enhanced Fingerprinting Based Indoor Positioning Using Machine Learning显示文摘Due to the inability of the Global Positioning System(GPS)signals to penetrate through surfaces like roofs,walls,and other objects in indoor environments,numerous alternative methods for user positioning have been presented.Amongst those,the Wi-Fi fingerprinting method has gained considerable interest in Indoor Positioning Systems(IPS)as the need for lineof-sight measurements is minimal,and it achieves better efficiency in even complex indoor environments.Offline and online are the two phases of the fingerprinting method.Many researchers have highlighted the problems in the offline phase as it deals with huge datasets and validation of Fingerprints without pre-processing of data becomes a concern.Machine learning is used for the model training in the offline phase while the locations are estimated in the online phase.Many researchers have considered the concerns in the offline phase as it deals with huge datasets and validation of Fingerprints becomes an issue.Machine learning algorithms are a natural solution for winnowing through large datasets and determining the significant fragments of information for localization,creating precise models to predict an indoor location.Large training sets are a key for obtaining better results in machine learning problems.Therefore,an existing WLAN fingerprinting-based multistory building location database has been used with 21049 samples including 19938 training and 1111 testing samples.The proposed model consists of mean and median filtering as pre-processing techniques applied to the database for enhancing the accuracy by mitigating the impact of environmental dispersion and investigated machine learning algorithms(kNN,WkNN,FSkNN,and SVM)for estimating the location.The proposed SVM with median filtering algorithm gives a reduced mean positioning error of 0.7959 m and an improved efficiency of 92.84%as compared to all variants of the proposed method for 108703 m^(2) area. | Muhammad Waleed Pasha Mir Yasir Umair Alina Mirza Faizan Rao Abdul Wakeel Safia Akram Fazli Subhan Wazir Zada Khan | 2021 | Computers, Materials & Continua2021，，11： | 0 |
| 16 | Anisotropic phonon thermal transport in two-dimensional layered materials显示文摘Two-dimensional layered materials(2DLMs)have attracted growing attention in optoelectronic devices due to their intriguing anisotropic physical properties.Different members of 2DLMs exhibit unique anisotropic electrical,optical,and thermal properties,fundamentally related to their crystal structure.Among them,directional heat transfer plays a vital role in the thermal management of electronic devices.Here,we use density functional theory calculations to investigate the thermal transport properties of representative layered materials:β-InSe,γ-InSe,MoS2,and h-BN.We found that the lattice thermal conductivities ofβ-InSe,γ-InSe,MoS_(2),and h-BN display diverse anisotropic behaviors with anisotropy ratios of 10.4,9.4,64.9,and 107.7,respectively.The analysis of the phonon modes further indicates that the phonon group velocity is responsible for the anisotropy of thermal transport.Furthermore,the low lattice thermal conductivity of the layered InSe mainly comes from low phonon group velocity and atomic masses.Our findings provide a fundamental physical understanding of the anisotropic thermal transport in layered materials.We hope this study could inspire the advancement of 2DLMs thermal management applications in next-generation integrated electronic and optoelectronic devices. | Yuxin Cai Muhammad Faizan Huimin Mu Yilin Zhang Hongshuai Zou Hong Jian Zhao Yuhao Fu Lijun Zhang | 2023 | Frontiers of physics2023，18，4： | 0 |
| 17 | 空位有序双钙钛矿A2BX6的弹性和热电性质的第一性原理研究显示文摘卤化物钙钛矿在热电应用中表现出了巨大的潜力。准确了解卤化物钙钛矿的热电传输性质对于进一步提高热电设备的应用效率至关重要。本研究采用了Perdew-Burke-Ernzerhof(PBE)和修正的Becke Johnson(mBJ)交换关联泛函探究了卤化物双钙钛矿Rb_(2)SnI6、Rb_(2)PdI6和Cs_(2)PtI6的弹性和热电性质。通过对这些材料的力学稳定性、有效质量、塞贝克系数、功率因子和热电品质因数的研究,我们发现,这三种化合物都是力学稳定的,并且具有可塑性。这些化合物是窄带隙半导体,具有简并的带边结构,结合低的载流子有效质量使得它们具有热电应用潜力。Cs_(2)PtI6在空穴掺杂在500 K温度下可以达到0.76 mV·K^(−1)的高塞贝克系数。由于高塞贝克系数和最大功率因子,Rb_(2)SnI6、Rb_(2)PdI6和Cs_(2)PtI6在p型掺杂下具有高的热电品质因数,分别为0.91、0.96和0.98。总体而言,我们的研究为卤化物钙钛矿的热电性能提供了新的见解,为这些化合物的实验合成提供了有价值的参考。 | Muhammad Faizan 赵国琪 张天旭 王啸宇 贺欣 张立军 | 2024 | 物理化学学报2024，40，1： | 0 |
| 18 | Large Scale Fish Images Classification and Localization using Transfer Learning and Localization Aware CNN Architecture显示文摘Building an automatic fish recognition and detection system for largescale fish classes is helpful for marine researchers and marine scientists because there are large numbers of fish species.However,it is quite difficult to build such systems owing to the lack of data imbalance problems and large number of classes.To solve these issues,we propose a transfer learning-based technique in which we use Efficient-Net,which is pre-trained on ImageNet dataset and fine-tuned on QuT Fish Database,which is a large scale dataset.Furthermore,prior to the activation layer,we use Global Average Pooling(GAP)instead of dense layer with the aim of averaging the results of predictions along with having more information compared to the dense layer.To check the validity of our model,we validate our model on the validation set which achieves satisfactory results.Also,for the localization task,we propose an architecture that consists of localization aware block,which captures localization information for better prediction and residual connections to handle the over-fitting problem.Actually,the residual connections help the layer to combine missing information with the relevant one.In addition,we use class weights and Focal Loss(FL)to handle class imbalance problems along with reducing false predictions.Actually,class weights assign less weights to classes having fewer instances and large weights to classes having more number of instances.During the localization,the qualitative assessment shows that we achieve 57%Mean Intersection Over Union(IoU)on testing data,and the classification results show 75%precision,70%recall,78%accuracy and 74%F1-Score for 468 fish species. | Usman Ahmad Muhammad Junaid Ali Faizan Ahmed Khan Arfat Ahmad Khan ArifUr Rehman Malik Muhammad Ali Shahid Mohd Anul Haq Ilyas Khan Zamil SAlzamil Ahmed Alhussen | 2023 | Computer Systems Science & Engineering2023，45，5： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员