Please wait a minute...
, No 8

■ Cover page(PDF 3.15 MB) ■  Table of Content, August 2023(PDF 33 KB)  ■  Combined articles (PDF 3.86 MB)

  • Real-Time Crop Disease Detection and Remedial Suggestion through Deep Learning-based Smartphone Application
    Kavita Pandey, and Dhiraj Pandey
    2023, 19(8): 491-498.  doi:10.23940/ijpe.23.08.p1.491498
    Abstract    PDF (629KB)   
    References | Related Articles

    More than half of the workforce of many countries, such as India, are still engaged majorly in agriculture, according to a survey. Crop diseases are a major threat to food security that farmers grow every year. The early identification of crop disease remains difficult in many parts of India due to the lack of the necessary infrastructure. Several solutions have been devised at the governmental level to address the challenge of food security. Still, most Indian farmers do not have sufficient technical support to address major problems like monitoring fields, which includes irrigation control, soil moisture, invigilating water level, and detection of crop diseases. A solution in an affordable form that satisfies the Indian context is highly needed. In this article, the issue of crop disease detection has been addressed using the advanced technologies that can be provided in low-cost smartphones. Timely identification of diseases and subsequent immediate remedial action will help in saving the yields which automatically saves the economy of the farmer and in turn can help several farmers from distress. A deep learning-based real-time solution has been proposed that ensures ease of access, convenient architecture, and 24*7 connectivity by empowering the user with the element of Disease Prediction and Remedy suggestion.

    UWGAN-EnhaNet: Conditional Generative Adversarial Network Inspired Network for Enhancing the Quality of Underwater Images
    M. J. Delsey and J. V. Bibal Benifa
    2023, 19(8): 499-506.  doi:10.23940/ijpe.23.08.p2.499506
    Abstract    PDF (575KB)   
    References | Related Articles

    Researchers have been focusing on uncovering underwater treasures by overcoming the obstacles of poor quality underwater images.. The onset of deep learning methods and the various acquisitions of underwater images paved the way for a lot of explorations. In this paper, architecture based on GAN is proposed to enhance the characteristics of the underwater image by preserving the structure and content of the image. Experiments are executed by utilizing the publicly available UFO-120 and UIEB datasets which include both real undersea images and their corresponding reference images. To boost the performance of the architecture, ℒ1 and content-based loss are combined with ℒcGAN. The final enhanced image provides an appealing result in qualitative evaluation whereas the results obtained from PSNR, SSIM, and UIQM metrics demonstrate that the suggested strategy produces improved results when compared with the most recent techniques.

    Hybrid Machine Learning Model for Load Prediction in Cloud Environment
    Savita Khurana, Gaurav Sharma, and Bhawna Sharma
    2023, 19(8): 507-515.  doi:10.23940/ijpe.23.08.p3.507515
    Abstract    PDF (515KB)   
    References | Related Articles

    Virtual machine (VM) load prediction is a critical task in cloud computing. Accurate VM load prediction can help to improve resource utilization, reduce costs, and improve the quality of service. For the Hybrid LSTM and AdaBoost model, a novel approach is proposed for accurate VM load prediction in cloud environments. The proposed model combines the power of LSTM and AdaBoost model, aiming to capture temporal dependencies in the VM load data and enhance prediction accuracy. The proposed model leverages LSTM to learn patterns and dynamics from historical load data, while AdaBoost is used to create an ensemble of weak regressors that collectively make load predictions. The model follows a two-step process: first, LSTM is trained on historical load data to extract informative features, and then AdaBoost is trained to combine the predictions from multiple weak regressors. The hybrid model demonstrates improved performance in VM load prediction by effectively handling non-linear relationships, temporal dependencies, and complex load patterns. The outcome of the proposed model is calculated using metrics, i.e. MAE, MAPE, MSE, RMSE, R2, and compared with existing machine learning algorithms i.e., Adaboost, KNN, SVM and deep learning algorithms i.e. LSTM, RNN. The results clearly show the superiority of the proposed hybrid approach in accurately predicting virtual machine load, enabling efficient resource allocation and management in cloud computing environments.

    Influence of Lean and Lean Six Sigma on Social Factors in the Moroccan Industry - Case Study
    Abdelouahed Boutayeb, Abderrahim Chamat, Abdelali Ennadi, and Abdelmajid Daya
    2023, 19(8): 516-525.  doi:10.23940/ijpe.23.08.p4.516525
    Abstract    PDF (574KB)   
    References | Related Articles

    After World War II, the market experienced significant growth, leading to a shift in demand compared to supply. Moreover, the increase in purchasing power resulted in a massive surge in consumption. In order to meet these demands, companies need to satisfy customers in terms of diversity, quality, price, and delivery time. To achieve this, they employ continuous improvement tools such as Lean and Lean Six Sigma (LSS). The primary goal of continuous improvement is to enhance performance in terms of quality, costs, and delivery time. However, these approaches often have negative consequences on social factors such as safety and ergonomics (SE), since they primarily aim to reduce waste and improve economic efficiency. In this article, we investigate the impact of Lean and LSS on SE factors using a case study. We also introduce our new concept of continuous improvement that considers SE social factors. This concept aims to enhance economic performance while ensuring the safety and well-being of workers.

    Adaptive Approach for Dynamic Spectrum Utilization in Wireless Communication System
    Niranjan S. Kulkarni, Sanjay L. Nalbalwar, and Anil B. Nandgaonkar
    2023, 19(8): 526-535.  doi:10.23940/ijpe.23.08.p5.526535
    Abstract    PDF (520KB)   
    References | Related Articles

    The evolution of wireless communication with new architectures has led to faster and more reliable wireless communication. However, with the rapid increase in the demanded services and user interface, available resources are becoming constraints in providing high-Quality of Services (QoS). Very High-Frequency Land Mobile Radio (LMR) communication is used as a means of data exchange over long-range wireless communication. LMR is designed with spectrum-sharing capability for various resource constraint applications. The existing approach of spectrum sensing using an energy-based detection technique is the widely used method in spectrum sensing and allocation. Various previous methods defined for spectrum utilization are developed with an assumption of a linear varying channel model, however, the dynamic variation in user interface and varying channel interference develops a limitation in resource utilization under dynamic channel conditions. Learning methods developed in optimizing resource allocation observe a large processing overhead under dynamic conditions. Addressing the issue of dynamic communication conditions, this paper outlines a method for Adaptive learning of estimation parameters in spectrum sharing for the cognitive wireless communication system. The dynamic spectrum variation is monitored in the resource-sharing process for higher system performance. The observations for the developed method illustrate an increase in system throughput, less delay, and system overhead in the network.

    An Advanced Machine Learning Approach for Student Placement Prediction and Analysis
    K. Eswara Rao, Bala Murali Pydi, T. Panduranga Vital, P. Annan Naidu, U. D. Prasann, and T. Ravikumar
    2023, 19(8): 536-546.  doi:10.23940/ijpe.23.08.p6.536546
    Abstract    PDF (905KB)   
    References | Related Articles

    As there are job opportunities worldwide, the graduates who are being produced in large numbers from various backgrounds are constantly trying to get them. Moreover, the management of graduate colleges gives proper training to the students to get those opportunities. Every student has their skills, unique creative outlook, studying, and good academic skills that help them get placed in various companies and also have a chance to get reputed positions, but most of the graduates are still failing to get the opportunity because they cannot find what skills to acquire. For this reason, in this paper, we gathered information from students who have finished their courses at different colleges. Collected information by communication and asked them about their academics, performance, families, skills, personal information, habits, etc., and what prevented them from taking the opportunity. Then, we made a dataset with all the factors that affected a student's career and used that to create a model with synthetic data. Student Placement Prediction can also benefit colleges and universities by providing valuable observations of student career outcomes. By understanding the factors influencing student job placement, colleges can conduct services and programs to help their students be better prepared for their careers. Accuracy and precision were used to evaluate the eXtreme Gradient Boost (XGBoost) machine learning model's performance compared to standard classification techniques. According to the results, the proposed algorithm is vastly superior to the alternatives.

    Enhancing Deception Detection with Exclusive Visual Features using Deep Learning
    Victor Diaz, W. Eric Wong, and Zizhao Chen
    2023, 19(8): 547-558.  doi:10.23940/ijpe.23.08.p7.547558
    Abstract    PDF (480KB)   
    References | Related Articles

    A combination of nonverbal cues, verbal cues, and measurements of body abnormality make guidelines to determine deceitfulness. The combination of these guidelines will vary from person to person, making deception detection a complex challenge. Research has demonstrated that the accuracy of the latest computerized polygraph testing techniques is 98% accurate. Several human-controlled variables help to achieve this level of accuracy, such as being properly trained and must use an accepted procedure and scoring system from the British Polygraph Society. This causes a lack of availability for Deception detection as the implementing these techniques have training from the British Polygraph Society. Hence this research aims to reduce the requirements of lie detection by relying on Visual Features tracked with computer vision. The proposed multi-modal will track facial and body movements to classify whether a person is Deceiving or telling the Truth. The model proposed will use data consisting of videos collected from public court trials. The data will be cleaned with Facial Action Units (AU) with OpenFace, and then augmented with various rotations. The features extracted from the videos are the Movement with Holistic landmarks and Unique features from deep learning extraction. The Multi-model will consist of three pathways: a 3D-CNN pathway, a CovLSTM2D Pathway, and a dense pathway. The outputs of the three paths are concatenated and fed into a dense layer with SoftMax activation for classification. With a continuous emphasis on examining the proposed methodology for model creation, we discovered that higher accuracy can be achieved by leveraging deep learning algorithms for visual inputs as complex as the human body.

ISSN 0973-1318