BengaliNet: A Low-Cost Novel Convolutional Neural Network for Bengali Handwritten Characters Recognition

Sayeed, Abu; Shin, Jungpil; Hasan, Md. Al Mehedi; Srizon, Azmain Yakin; Hasan, Md. Mehedi

doi:10.3390/app11156845

Cited by 23 publications

(6 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Earlier research on Bengali handwritten digit recognition relied on conventional machine learning techniques such as k-nearest neighbors (KNN) [6] and support vector machines (SVM) [7]. BengaliNet, a cost-effective convolutional neural network architecture was presented for recognizing Bengali characters [8].…”

Section: Previous Research On Bengali Handwritten Digit and Mathemati...mentioning

confidence: 99%

WITHDRAWN: A* Path Finding Algorithms Based Designing the Best Possible Eco-Friendly Structure Spatial Landscape and Natural Setting.

Imran

Das

Galib

et al. 2023

Preprint

View full text Add to dashboard Cite

The design of conventional architectural landscapes often has issues with inadequate direct visual effects and inadequate levels of resident satisfaction, but the design of green architectural landscapes typically does not have these issues. Creating room for landscaping is likewise an impractical endeavor. Because of this, the authors of this study suggest a number of different Path-finding algorithms for green building landscape space and environmental optimization design. A variety of path-finding algorithms may be used to maximize the multi-objective scale of the architectural landscape. The expense of the construction as well as public utilities such as the space environment are included among the optimization's goals and objectives. The findings indicate that the many pathfinding solutions that were presented are able to optimize the architectural landscape setting, and the occupants' level of happiness has increased as a consequence.

show abstract

Section: Previous Research On Bengali Handwritten Digit and Mathemati...mentioning

confidence: 99%

WITHDRAWN: A* Path Finding Algorithms Based Designing the Best Possible Eco-Friendly Structure Spatial Landscape and Natural Setting.

Imran

Das

Galib

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…A low-cost convolutional neural network design for Bengali handwritten character identification is suggested by the writers in the study [4]. For the training phase, the researchers used openly accessible standard datasets, such as the CMATERdb Bengali handwritten character dataset, and created various dataset forms in accordance with prior research.…”

Section: Previous Workmentioning

confidence: 99%

An Artificial Neural Network Method for Managing Inventory of A Fertilizer Company in Bangladesh

Hasan

Akhter

2022

Dhaka Univ. J. Sci.

View full text Add to dashboard Cite

Business organizations are always facing uncertainties in demand, supply and inventories. For this, it is important for them to make the strategic plans to cope up with the uncertainty accordingly. To sustain, the business organizations must plan the future in such a way that the inventory cost, labor cost will be minimized and the utilization of time, financial resources and profit will be maximized. The optimum planning of resources also help the organizations to avoid wastage. A good forecasting technique can help the manager of a company to deal with the uncertainties. In this paper, we will work on such a planning fora fertilizer company in Bangladesh. To minimize the inventory cost, we will apply a new approach known as Artificial Neural Network (ANN), which is recently used for the problem of prediction and analyze the main characteristics of a system through an iterative training process. For this, we will first forecast the demand of fertilizer by using existing forecasting methods. We will then apply ANN for forecasting the demand of fertilizer of the company. We will also identify Economic Order Quantity (EOQ) to minimize total cost including inventory costs of the company. We use programming language MATLAB for analyzing different forecasting methods including ANN. Finally, we will use these results to find out the right forecasting technique for the fertilizer company with optimal inventory cost. Dhaka Univ. J. Sci. 69(3): 133-142, 2022 (June)

show abstract

“…It won ILSVRC 2015 [12]. MobileNet optimizes CNNs for mobile/embedded devices via depthwise separable convolutions [13]. It splits convolutions into depthwise and pointwise steps, enhancing efficiency [14].…”

Section: Introductionmentioning

confidence: 99%

KurdSet Handwritten Digits Recognition Based on Different Convolutional Neural Networks Models

Hasen Ali,

Abdulrazzaq

2024

TEM Journal

View full text Add to dashboard Cite

Recognition of handwritten digits has garnered significant interest among researchers in the domain of recognizing pattern. This interest stems from the recognition's relevance in various real-life applications, including reading financial checks and official documents, which has remained a persistent obstacle. To address this challenge, researchers have developed numerous algorithms focusing on recognizing handwritten digits across different human languages. This paper presents a new Kurdish Handwritten dataset, consisting of Kurdish characters, digits, texts, and symbols. The dataset consists of 1560 participants, encompassing a broad and varied group. It serves as the primary dataset for training and evaluating algorithms in Kurdish digit recognition. We used Kurdish dataset named (KurdSet) and Arabic dataset for handwritten recognition, which holds 70,000 images of Arabic digits that were written by 700 various participants. Additionally, various models are utilized in the study, including ResNet50, DenseNet121, MobileNet, and a custom CNN (convolutional neural network). Additionally, the models' effectiveness was assessed through the examination of test accuracy, which measures the percentage of correctly classified digits in the evaluation phase. ResNet50 also performs exceptionally well that achieved test accuracy 99.67%, indicating its All models exhibit good performance, DenseNet121 and the Custom CNN Model demonstrate the highest test accuracy of 99.73%, highlighting their superior performance. capabilities in capturing relevant features. Despite its accuracy, MobileNet still exhibits good recognition capability with a test accuracy 99.54%.

show abstract

BengaliNet: A Low-Cost Novel Convolutional Neural Network for Bengali Handwritten Characters Recognition

Cited by 23 publications

References 45 publications

WITHDRAWN: A* Path Finding Algorithms Based Designing the Best Possible Eco-Friendly Structure Spatial Landscape and Natural Setting.

WITHDRAWN: A* Path Finding Algorithms Based Designing the Best Possible Eco-Friendly Structure Spatial Landscape and Natural Setting.

An Artificial Neural Network Method for Managing Inventory of A Fertilizer Company in Bangladesh

KurdSet Handwritten Digits Recognition Based on Different Convolutional Neural Networks Models

Contact Info

Product

Resources

About