Shakil Ahmed

Mr. Shakil Ahmed

American International University-Bangladesh
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 December 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Shakil AhmedShakil Ahmed
Shakil Ahmed
ASA F M Saifuddin Saif
A F M Saifuddin Saif
IHImtiaz Hanif
Imtiaz Hanif
MSMostofa Nurannabi Shakil
Mostofa Nurannabi Shakil
MJMostofa Jaman
Mostofa Jaman
HSHasan Sabbir
Hasan Sabbir
With the advancement of the technological field, day by day, people from around the world are having easier access to internet abled devices, and as a result, video data is growing rapidly. The increase of portable devices such as various action cameras, mobile cameras, motion cameras, etc., can also be considered for the faster growth of video data. Data from these multiple sources need more maintenance to process for various usages according to the needs. By considering these enormous amounts of video data, it cannot be navigated fully by the end-users. Throughout recent times, many research works have been done to generate descriptions from the images or visual scene recordings to address the mentioned issue. This description generation, also known as video captioning, is more complex than single image captioning. Various advanced neural networks have been used in various studies to perform video captioning. In this paper, we propose an attention-based Bi-LSTM and sequential LSTM (Att-BiL-SL) encoder-decoder model for describing the video in textual format. The model consists of two-layer attention-based bi-LSTM and one-layer sequential LSTM for video captioning. The model also extracts the universal and native temporal features from the video frames for smooth sentence generation from optical frames. This paper includes the word embedding with a soft attention mechanism and a beam search optimization algorithm to generate qualitative results. It is found that the architecture proposed in this paper performs better than various existing state of the art models.

1 January 2020 in International Journal of Advanced Computer Science and Applications
Verified
Journal Article
Open access
PDF file
IHImtiaz Hanif
Imtiaz Hanif
Shakil AhmedShakil Ahmed
Shakil Ahmed
WAWahiduzzaman Akanda
Wahiduzzaman Akanda
SBShohag Barman
Shohag Barman
Now-a-days, the public, mostly women and children are facing much harassment from the societies. The unlawful activities against ladies and children have been increasing signifi-cantly, and regularly we find out about eve-teasing, sexual assault cases, and attempt to molest or even killing after rape in public places or open areas. Also, many cases had gone unwarranted due to short pieces of evidence. In Bangladesh, the current statistics of sexual assaults and various unlawful activities are proliferating. To acknowledge these problems, in this paper, we have designed an IoT-based (Internet of Things) embedded device that is able to communicate with the law enforcement agency by dialing “999” (An Emergency Telephone Number in Bangladesh) on demand. The device contains Arduino Pro-Mini Microcontroller with a GSM (Global System for Mobile communication) module and can send SMS (short message service) with the victim’s present area to the law enforcement agency and relatives via GPRS (General Packet Radio Services). The proposed device’s form factor is too tiny to carry out easily at anywhere and anytime. The device features the “Plug & Play” functionalities, which means one button to operate the entire device. Also, the device is cost-effective so that people of every level can afford it at a reasonable price.

1 February 2020
Verified
Conference Paper
Closed access
Shakil AhmedShakil Ahmed
Shakil Ahmed
FKFazle Rabbi Khan
Fazle Rabbi Khan
FLFarhan Anjum Labib
Farhan Anjum Labib
ACAzm Ehtesham Chowdhury
Azm Ehtesham Chowdhury
The detection of an object or moving object is a comprehensive goal in the research domain in the era of computer vision. Numerous scientists are quick to think about the underwater world and the living articles in the deep ocean and they are incompletely effective to their objective. In this research work, , we observe the potentiality in underwater by a vision based object detection sphere by applying various previous work for highlighting the arising consequences in various underwater situation.. We have collected the dataset from various source of water such as clean water, turbid water, bubbles water etc. We used OpenCV and Python programming language to detect the object. Finally the detection of object in underwater is studied for connecting with the basic color based HSV(Hue, Saturation, Value) segmentation.

16 October 2026 in bioRxiv
Verified
Preprint
Open access
PDF file
Shakil AhmedShakil Ahmed
Shakil Ahmed
Kim BjergeKim Bjerge
Kim Bjerge
Toke HoyeToke Hoye
Toke Hoye
CPChristian Nørgaard Storm Pedersen
Christian Nørgaard Storm Pedersen
The increasing automation of biodiversity monitoring is generating large volumes of ecological images requiring taxonomic identification, creating a substantial annotation bottleneck. Assigning species-level labels requires considerable time and expert knowledge, limiting scalability. We propose BioCluster, a framework for unsupervised clustering and zero-shot labeling of ecological images using foundation models. The framework extracts image embeddings from pretrained models and groups images into taxonomic clusters without task-specific training. Species labels are assigned through zero-shot text matching against the taxa expected in the dataset. We investigate dimensionality reduction and six clustering algorithms for organizing fine-grained imagery. A per-image uncertainty score combines clustering and labeling information to sort images into three confidence tiers, so experts can focus on uncertain and unlabeled cases. We evaluate it on two moth and two bird datasets, with one annotated camera-trap collection per taxonomic group. For these camera-trap datasets, per-image species-level accuracy reached 84% on DK-Moth and 83% on HUGO-Bench Aves. An interactive report presents clusters and zero-shot labels by confidence tier and prioritizes cases requiring expert review. The framework provides a training-free approach for organizing and labeling ecological image collections, reducing manual annotation while retaining expert oversight and demonstrating the potential of foundation models for scalable, semi-automated annotation.

24 October 2024
Verified
Conference Paper
Closed access
MHMazid- Ul- Haque
Mazid- Ul- Haque
Shakil AhmedShakil Ahmed
Shakil Ahmed
Rakin Sad AftabRakin Sad Aftab
Rakin Sad Aftab
SMSaef Ullah Miah
Saef Ullah Miah
WAWahiduzzaman Akanda
Wahiduzzaman Akanda
ABAbhijit Bhowmik
Abhijit Bhowmik
The Internet is integral to modern life, with Internet service provider (ISP) offering appealing deals to meet the demand for unlimited data. However, reality often falls short of expectations. While recommendation systems exist, user-centric options are rare. This paper proposes a novel ISP selection methodology using user experience data and a calibrated random forest (CRF) model. Unlike traditional methods that focus on advertised features, this approach emphasizes user-defined criteria such as cost, device connectivity, and technical support experience. By analyzing survey data, the model highlights the critical link between user needs and support quality, enabling users to choose ISPs that prioritize customer service. The model demonstrates promising results with a strong R-squared value and low mean squared error (MSE). This user-centric approach fosters informed decision-making, potentially driving competition and encouraging ISPs to improve service standards, laying a foundation for future developments in ISP selection.