Human gait recognition (HGR) has been employed as a biometric technique for security purposes over the last decade. Various factors, including clothing, carrying items, and walking surfaces, can influence the performance of gait recognition. Additionally, identifying individuals from different viewpoints presents a significant challenge in HGR. Numerous conventional and deep learning techniques have been introduced in the literature for HGR, but traditional methods are not well suited to handling large datasets. This research explores the effectiveness of four deep learning models for gait identification in the CASIA B dataset: the convolutional neural network (CNN), multi-layer perceptron (MLP), self-organizing map (SOMs), and transfer learning with EfficientNet. The selected deep learning techniques offer robust feature extraction and the efficient handling of large datasets, making them ideal in enhancing the accuracy of gait recognition. The collection includes gait sequences from 10 individuals, with a total of 92,596 images that have been reduced to 64 × 64 pixels for uniformity. A modified model was developed by integrating sequential convolutional layers for detailed spatial feature extraction, followed by dense layers for classification, optimized through rigorous hyperparameter tuning and regularization techniques, resulting in an accuracy of 97.12% for the test set. This work enhances our understanding of deep learning methods in gait analysis, offering significant insights for choosing optimal models in security and surveillance applications.

Gait recognition is an important specific type of video-based behavior analysis. Such methods are potentially important for any type of video surveillance where specific people are being tracked or where the movement behavior of certain people is being analyzed because people's movements have unique characteristics. For example, investigating an individual without a representative of dangerous or hostile behavior. As a result, the gait feature can be used to capture long distances and uncontrolled situations more easily. Therefore, gait can be used as a biometric measure to identify known individuals and classify subjects. To this end, we designed a human gait classifier that facilitates human recognition through their walk. In this paper, we first obtain silhouette images from the CASIA B dataset. After that, we create 300 sequences by aspect ratio carve and local minima for 10 subjects and divide them into two directories named train and test. The proposed method used the CNN pre-trained xception model to extract features from the sequences. The given extracted frames are then trained and classified using the RNN model. Finally, we evaluated the performance using a hybrid neural network (CNN-RNN) which gave 90.00% accuracy to the trained model.

Rice is the most widely consumed staple food of Bangladesh. Rice plants are extremely affected by the primary nutrient shortage elements like nitrogen, phosphorus, and potassium. The rice leaves are influenced by the deficiency of primary nutrient ingredients, simultaneously. The unhealthy rice leaves are dominated by several colors and patterns. The proposed work in this presented paper aims to identify and classify the nitrogen, phosphorus, and potassium nutrient deficiency in rice leaves image. The experiment is conducted with a k-means clustering algorithm and convolutional neural network. Firstly, the rice leaf images are segmented with multiple K values by the k-means clustering algorithm so that defective leaves are accurately recognized for classification. Secondly, the different segmented images are employed by the classifier algorithm named convolutional neural network. As a result, the various K cluster value gives the accuracy that is 81%, 83%, 88%, 90%, 85%, 85%, and 79% respectively. Finally, the results are compared with respect to the K cluster value and extracted the standard K value from the k-means cluster numbers.

Gait recognition, a form of biometric identification, has garnered considerable interest for its ability to identify individuals from a distance. Yet, its effectiveness depends significantly on the precision of the underlying categorization models. This study thoroughly evaluates many Convolutional Neural Network (CNN) models to determine their effectiveness in properly identifying and categorizing gait patterns. We train and test several CNN architectures using advanced deep learning techniques on the widely known CASIA-B dataset, which is a benchmark dataset for gait recognition research. The analyses involve models like as VGG16, VGG19, NASNetLarge, NASNetMobile, EfficientNetB0, and Xception, each well-known for their effectiveness in picture categorization tasks. We evaluate the performance of these models by conducting thorough experiments and detailed analysis, focusing on accuracy, validation loss, and other pertinent metrics. The Xception model demonstrated the highest accuracy of 97.17% of the models assessed. This model frequently surpasses similar models, demonstrating its strength and effectiveness in precisely recognizing gait patterns. On the other hand, the accuracy of other models varied from 9.78% to 93.85%, demonstrating the variable levels of efficacy among various designs. The findings have significant implications for the creation and implementation of gait recognition systems. The high precision demonstrated by the Xception model highlights the promise of CNN-based methods in progressing gait identification technology. Our work highlights the significance of choosing suitable model architectures for achieving the best performance in gait detection tasks. In the future, research may focus on using larger datasets and exploring other CNN architectures to improve the accuracy and reliability of gait detection systems. Our goal is to enhance gait recognition technology to improve biometric identification for security, surveillance, and healthcare applications. GUB JOURNAL OF SCIENCE AND ENGINEERING, Vol 10(1), 2023 P 17-26

Acute Lymphoblastic Leukemia (ALL) is an aggressive blood cancer that affects white blood cells; early and accurate diagnosis is essential for the institution of proper treatment. Traditional methods of leukemia diagnosis, through manual investigation of blood smears, are time-consuming and prone to human error, hence there is a need for more robust automated solutions for diagnosis. This study presents a novel hybrid deep learning model by combining the Xception and ResNet50 models in endeavoring to address the challenges of feature extraction and classification accuracy in microscopic blood smear images. Using convolutional neural networks with some optimized preprocessing, such as image cropping, normalization, and augmentation of data, this study presents a model that can achieve very exceptional diagnostic performance in classifying cancer, at an accuracy of 99.99% on the Kaggle ALL dataset, outperforming other models like ResNet34 and GoogLeNet. Exhaustive testing shows strong generalization with minimal loss and low overfitting, which demonstrates the robustness of the model and its efficiency in picking up subtle patterns at a cell-specific level. This research highlights the transformative potential of hybrid deep learning models in clinical diagnostics for more accurate diagnosis, efficiency, and integration into clinical workflows of early Acute Lymphoblastic Leukemia (ALL) detection for the benefit of patient outcomes and the advancement of research on automated leukemia diagnostics.

In computer vision, accurately identifying pedestrians is a crucial difficulty, especially in applications where safety is of the highest priority, such autonomous driving and urban planning. Deep learning has made progress, but challenges such changing conditions, barriers, and processing power requirements still exist. In this paper, we present an improved deep convolutional neural network (DCNN) designed with pedestrian identification in mind. Improving accuracy in various settings while maximizing processing efficiency is our goal. We carefully prepare the data and create a novel DCNN architecture that can handle a variety of pedestrian appearances and urban surroundings by utilizing well-known datasets such as Penn-Fudan and PnPLO. Our model performs exceptionally well in terms of accuracy, speed, and durability, according to a thorough evaluation that compares it to recognized benchmarks. In addition to making important advances, our study has shown novel approaches to enhancing performance in challenging environments, such as the analysis of transfer learning and the deployment of real-time systems. This work paves the way for future developments in intelligent, self-governing systems and is a maj or leap forward in pedestrian detection technology.