Deep Neural Network Algorithm Research Articles

Genetic and behavioural factors affecting interpopulation colour pattern variation in two congeneric chameleon species.

We conducted a study on interpopulation variation of colour patterns in two congeneric chameleon species, which have an analogous life history. Both species are able to rapidly change colour pattern, and their context-dependent colour patterns often vary across a wide geographical range. Specifically, we tested four hypotheses that can explain the observed interpopulation variation of colour patterns by a series of behavioural field trials where the colour patterns of individuals were recorded and later analysed by a deep neural network algorithm. We used redundancy analysis to relate genetic, spectral and behavioural predictors to interpopulation colour pattern distance. Our results showed that both isolation by distance (IBD) and alternative mating tactics were significant predictors for interpopulation colour pattern variation in Chamaeleo chamaeleon males. By contrast, in Chamaeleo dilepis, the interpopulation colour pattern variation was largely explained by IBD, and evidence for alternative mating tactics was absent. In both chameleon species, the environmental colours showed no evidence of influencing chameleon interpopulation colour pattern variation, regardless of sex or behavioural context. This contrasting finding suggests that interpopulation context-dependent colour pattern variations in each species are maintained under a different set of selective pressures or circumstances.

Design of Intelligent Sentiment Classification Model Based on Deep Neural Network Algorithm in Social Media

Design of Intelligent Sentiment Classification Model Based on Deep Neural Network Algorithm in Social Media

Evaluation of deep learning and convolutional neural network algorithms for mandibular fracture detection using radiographic images: A systematic review and meta-analysis

Evaluation of deep learning and convolutional neural network algorithms for mandibular fracture detection using radiographic images: A systematic review and meta-analysis

Automatic fabrication system of optical micro-nanofiber based on deep learning

<sec>The wide range, high precision, and dynamic real-time measurement of micro-/nanofiber diameter are crucial for achieving low loss transmission and controlling dispersion in the preparation process of micro-nanofiber. In view of the problems of small diameter regulation range, complex operation and long-time consumption of the existing preparation methods, the automatic detection system of micro-nanofiber is realized based on deep learning neural network algorithm in this work. The image segmentation method in computer vision is used to make high-quality multi-scale micro-/nanofiber datasets, and the improved YOLOv8-FD (You Only Look Once version 8-Fiber Detection) algorithm based on small target detection is used to automatically detect the diameter of micro-nanofiber.</sec><sec>Through image segmentation and identification of the target of single pixel size in the microscopic image, the diameter detection of micro-nanofiber is finally realized. In this process, the real-time diameter of micro-nanofiber is obtained through image information, and then the micro-nanofiber small target is accurately segmented to achieve the precise detection of mAP<sup>IoU</sup><sup>=50</sup> = 0.975 and mAP<sup>IoU</sup><sup>=50—95</sup> = 0.765 on the micro-nanofiber multi-scale target dataset with extremely high accuracy. The algorithm-based construction of a high-precision micro-nanofiber automatic preparation system enables real-time accurate segmentation of fiber edges, calculation of fiber diameter, and feedback to the control system for achieving automated preparation of fibers with arbitrary diameters. Additionally, it facilitates the detection of micro-nanofiber in a range from 462 nm to 125 μm. The average response time for reasoning is 9.6 ms, and the detection error is kept below 2.95%.</sec><sec>In addition, compared with other micro-/nanofiber diameter detection methods based on optical imaging and mode cutoff, this method shows advantages of high precision, high speed and arbitrary diameter preparation for diameter detection based on deep learning neural networks. The system is very suitable for high-precision real-time measurement and automatic and accurate preparation of micro-/nanofibers, thereby providing a novel method of developing micro-nanofiber devices with low-loss transmission and adjustable dispersion. </sec>

A novel rice plant leaf diseases detection using deep spectral generative adversarial neural network

The farming industry widely requires automatic detection and analysis of rice diseases to avoid wasting financial and other resources, reduce yield loss, improve processing efficiency, and obtain healthy crop yields. The proposed Deep Spectral Generative Adversarial Neural Network (DSGAN2) method is used for detecting rice plant leaf disease. Initially, fed into the input of healthy and non-healthy leaves from the collected dataset. Then, apply an Improved Threshold Neural Network (ITNN) method to enhance the image quality. Next, it uses a Segmentation using a Segment Multiscale Neural Slicing (SMNS) algorithm to identify the support-intensive color saturation based on the enhanced image. After that, the Spectral Scaled Absolute Feature Selection (S2AFS) method is applied to select optimal features and the closest weight from segmented rice plant leaves. Social Spider Optimization will select the feature using the Closest Weight (S2O-FCW) algorithm to analyze the feature weight values. Finally, the proposed Soft-Max Logistic Activation Function with Deep Spectral Generative Adversarial Neural Network (DSGAN2) algorithm detects rice plant disease based on selected features. With an accuracy of 97 %, the model helps farmers identify and identify Rice Plant diseases. The proposed system Deep Spectral Generative Adversarial Neural Network (DSGAN2) produces a decreasing false rate compared to the existing system of ACPSOSVM-Dual Channels Convolutional Neural Network (APS-DCCNN) is 55.2 %, Alex Net is 50.4 %, and Convolutional Neural Network (CNN) is 49.5 %.

Study on the Application of Improved Deep Convolutional Neural Network Algorithm in Broken Information Recovery

Abstract This study on the fusion of deep convolutional neural network (CNN) and extended short-term memory network (LSTM) aims to improve the efficiency and accuracy of broken information recovery. The challenges faced by traditional information recovery techniques are addressed through improved algorithms. The research methodology includes constructing CNN models to automatically extract features and combining LSTM networks to process complex time-series data. We conducted a detailed experimental evaluation of the CNN-LSTM fusion algorithm, including recovery of different types of corrupted data, and compared it with other algorithms. The results show that the CNN-LSTM fusion algorithm has the highest structural similarity (0.9545) and the most minor normalized mean square error (0.0016) for recovering broken video information, outperforming the methods using CNN or LSTM alone. The fusion algorithm dramatically reduces computation time and resource consumption for processing complex datasets. The combination of CNN and LSTM significantly improves the performance of broken information recovery, especially in processing video and audio data, and provides new ideas for future information processing techniques.

Application of Improved Deep Convolutional Neural Network Algorithm in Damaged Information Restoration

Application of Improved Deep Convolutional Neural Network Algorithm in Damaged Information Restoration

Enhancing Distributed Neural Network Training Through Node-Based Communications.

The amount of data needed to effectively train modern deep neural architectures has grown significantly, leading to increased computational requirements. These intensive computations are tackled by the combination of last generation computing resources, such as accelerators, or classic processing units. Nevertheless, gradient communication remains as the major bottleneck, hindering the efficiency notwithstanding the improvements in runtimes obtained through data parallelism strategies. Data parallelism involves all processes in a global exchange of potentially high amount of data, which may impede the achievement of the desired speedup and the elimination of noticeable delays or bottlenecks. As a result, communication latency issues pose a significant challenge that profoundly impacts the performance on distributed platforms. This research presents node-based optimization steps to significantly reduce the gradient exchange between model replicas whilst ensuring model convergence. The proposal serves as a versatile communication scheme, suitable for integration into a wide range of general-purpose deep neural network (DNN) algorithms. The optimization takes into consideration the specific location of each replica within the platform. To demonstrate the effectiveness, different neural network approaches and datasets with disjoint properties are used. In addition, multiple types of applications are considered to demonstrate the robustness and versatility of our proposal. The experimental results show a global training time reduction whilst slightly improving accuracy. Code: https://github.com/mhaut/eDNNcomm.

Intelligent Assessment of Teaching Quality of Computer Technology Courses in Colleges and Universities Based on ISSA-DRNN

Abstract With the continuous development of artificial intelligence technology, its application in the teaching of computer technology courses in colleges and universities is gradually deepened. In this study, the traditional sparrow search algorithm (SSA) is first processed to find the optimal solution, and then the ISSA algorithm is proposed and modeled. Training the deep recurrent neural network algorithm (DRNN) and combining it with the ISSA algorithm resulted in the construction of the teaching wisdom assessment model for the course. Finally, a teaching quality evaluation system for computer courses is constructed and applied to explore changes in teaching quality. The results show that in the assessment of the teaching quality of computer technology courses, the course structure and the teacher’s profile occupy the highest weight, and the weights occupied by these two items are more than 0.7. Secondly, the accuracy of the ISSADRNN model reaches more than 95%, and the maximum error is 0.0284. The average scores of each index system are more than 88. The overall performance is good and close to excellent. The model output’s actual value is within the expected range and has high accuracy. It can be seen that the intelligent assessment model of computer teaching quality proposed in this study has broad application potential and practical value.

Teaching Quality Assessment of English Courses in Colleges and Universities Based on ISSA-DRNN

Abstract The optimized sparrow search algorithm (ISSA) is built on top of the deep neural network (DNN) in this paper, along with a new deep recurrent neural network (DRNN) algorithm that is proposed by improving the DNN. Ultimately, an ISSA-DRNN-based model for assessing the quality of English course teaching has been established. To examine the application impact of the ISSA-DRNN-based English course quality teaching evaluation model, SA University has been chosen as the study location. Two groups have been created: an experimental group and a control group. The experimental group’s academic level scores grew by 5.1 points in terms of English achievement, whereas the control group’s average score increased by 0.36 points, indicating a very significant difference (P<0.01). The four characteristics of satisfaction with teaching effectiveness—interest in learning, ability improvement, desire to learn behavior, and course satisfaction— showed significant differences (P<0.05) among students in the experimental group. In the self-assessment of English literacy, the dimensions with the most important proportion of “relatively large” ratings are learning motivation and class participation, accounting for 43.73% and 42.92%, respectively. The dimensions with the highest proportion of “substantial” grades are English learning ability and learning motivation, accounting for 21.03% and 20.63% in that order.

Deep neural network with fuzzy algorithm to improve power and traffic-aware reliable reactive routing

<span>In wireless networks, link breaks, and restricted resources create fundamental challenges for maintaining network applications. Several wireless network routing techniques concentrate on power efficiency to expand the network lifetime, but the traffic and reliability parameters are not the primary concern. Though, these techniques are not capable of dealing with the wireless network. Hence, this paper proposes deep neural network (DNN) with a fuzzy algorithm to improve power and traffic-aware reliable reactive routing (PTAR) in wireless networks. The wireless network is formed by clustering by the node power and selects the cluster head (CH) based on a fuzzy algorithm. The wireless node power level, node buffer space, and node reliability to consider the input parameters of the fuzzy system. Then thefuzzy algorithm gives the output for CH round length. This selected CH improves the node reliability, power efficiency with minimized network congestion. Then we use a DNN algorithm to choose an optimal relay by applying an adaptive load balance factor in the network. DNN is a machine learning algorithm, and it provides high accuracy. From the simulation results, the PTAR approach improves the network performance, such as packet received ratio, delay, residual energy, and routing overhead.</span>

Predicting of a person's position in trajectory tracking from a continuous video stream

The paper proposes a method for predicting when a person enters a forbidden zone during his trajectory following a video stream, considering individual body parts. The authors used the PP-TinyPose PaddleHub neural network model with its implementation based on two deep neural networks to detect key points of the human body. The paper considers an example of human position prediction from a continuous video stream in indoor trajectory tracking. The authors predicted each key point in the coordinate space of the video stream using a recurrent deep neural network algorithm.

Business model innovation of advertising operation based on RBM deep neural network

Abstract RBM, as an important structural component unit in the mainstream model of deep learning, is being widely used in various fields. The article first discusses the RBM deep neural network algorithm, gives an overview of the related function algorithm combining RBM and neural network, and then proposes a deep network method oriented to the prediction of advertisement click rate. At the end of the article, the algorithm proposed in this paper is compared with the model performance test. An advertising company is used as an object of empirical research to analyze the change in its market share and turnover in the years 2022-2023. Explore the significance and role of RBM-based deep neural networks in the innovation of advertising business models.

A data-driven model for power system operating costs based on different types of wind power fluctuations

The stochastic and intermittent features of wind power as well as the high percentage of wind power grid-connected significantly increase the additional operating costs of the power system. It is difficult to accurately calculate the impact of complex fluctuations in wind power on additional operating costs. To solve the above problems, a power system operating cost model adapted to various wind power fluctuation processes is established. Firstly, based on a two-layer clustering strategy, different types of wind power fluctuations are obtained. Then, a production simulation model of the power system with renewable energy is established. The production simulation model costs include thermal plant operating costs, energy storage system operating costs, positive reserve costs and negative reserve costs. With the optimization objective of minimizing the total operating cost of the power system, realistic and representative system operating parameters and cost samples are obtained for various wind power fluctuations and different wind power grid-connected scenarios. Finally, a data-driven approach based on a deep neural network algorithm is proposed to achieve precise mapping between wind energy fluctuations and the operating costs of power systems and thermal power units, and the operating costs of the power system during the four seasons with different types of wind power fluctuations can be precisely analyzed. The results demonstrate that the method proposed in this paper has high simulation accuracy for the overall simulation operating cost of the power system and the operating cost of thermal power plants. The simulation errors are 4%–18% and 3%–13%, respectively, which verified the effectiveness of the method.

Optimization of energy consumption in smart homes using firefly algorithm and deep neural networks

Electronic gadget advancements have increased the demand for IoT-based smart homes as the number of connected devices grows rapidly. The most prevalent connected electronic devices are smart environments in houses, grids, structures, and metropolises. Smart grid technology advancements have enabled smart structures to cover every nanosecond of energy use. The problem with smart, intelligent operations is that they use a lot more energy than traditional ones. Because of the growing growth of smart cities and houses, there is an increasing demand for efficient resource management. Energy is a valuable resource with a high unit cost. Consequently, authors are endeavoring to decrease energy usage, specifically in smart urban areas, while simultaneously ensuring a consistent terrain. The objective of this study is to enhance energy efficiency in intelligent buildings for both homes and businesses. For the comfort indicator ("thermal, visual, and air quality"), three parameters are used: temperature, illumination, and CO2. A hybrid rule-based Deep Neural Network (DNN) and Fire Fly (FF) algorithm are used to read the sensor parameters and to operate the comfort indication, as well as optimize energy consumption, respectively. The anticipated user attributes contributed to the system's enhanced performance in terms of the ease of use of the smart system and its energy usage. When compared to traditional approaches in expressions of Multi View with 98.23%, convolutional neural network (CNN) with 99.17%, and traffic automatic vehicle (AV) with 98.14%, the activities of the contributed approach are negligibly commanding.

Brain age prediction using combined deep convolutional neural network and multi-layer perceptron algorithms

The clinical applications of brain age prediction have expanded, particularly in anticipating the onset and prognosis of various neurodegenerative diseases. In the current study, we proposed a deep learning algorithm that leverages brain structural imaging data and enhances prediction accuracy by integrating biological sex information. Our model for brain age prediction, built on deep neural networks, employed a dataset of 3004 healthy subjects aged 18 and above. The T1-weighted images were minimally preprocessed and analyzed using the convolutional neural network (CNN) algorithm. The categorical sex information was then incorporated using the multi-layer perceptron (MLP) algorithm. We trained and validated both a CNN-only algorithm (utilizing only brain structural imaging data), and a combined CNN-MLP algorithm (using both structural brain imaging data and sex information) for age prediction. By integrating sex information with T1-weighted imaging data, our proposed CNN-MLP algorithm outperformed not only the CNN-only algorithm but also established algorithms, such as brainageR, in prediction accuracy. Notably, this hybrid CNN-MLP algorithm effectively distinguished between mild cognitive impairment and Alzheimer’s disease groups by identifying variances in brain age gaps between them, highlighting the algorithm’s potential for clinical application. Overall, these results underscore the enhanced precision of the CNN-MLP algorithm in brain age prediction, achieved through the integration of sex information.

Heart Disease Classification Using Deep Neural Network with SMOTE Technique for Balancing Data

Heart disease is the leading cause of premature death worldwide. According to the WHO, heart disease causes about 30% of the total 58 million deaths and mostly occurs in individuals who are in their productive age. This condition can occur to anyone, including individuals who do not show symptoms of heart disease. However, heart disease can be prevented with early detection. By understanding the various risk factors that can increase the potential for heart disease. Therefore, this study aims to classify heart disease using Deep Neural Network algorithm and SMOTE technique to overcome data imbalance. This research resulted in a validation accuracy of 90% with precision evaluation of 0.85, recall 0.92, and f1-score 0.88. Based on the results obtained, the Deep Neural Network algorithm after SMOTE is superior to the model without SMOTE.

Moving scene object tracking method based on deep convolutional neural network

The effect of target tracking is not ideal when facing various complex tracking scenarios such as non-rigid deformation of target, frequent occlusion, clutter of target background and interference of similar objects. In this paper, the feature based on deep convolutional neural network is used for target tracking in moving scenes, and a sliding window target segmentation method is proposed to study the impact of data normalization and data set expansion on the final result. In order to select more distinguishing features, principal component analysis is used to process the features of Deep Convolution Neural Network (DCNN), and the features of different network layers of DCNN are compared. The feature coding algorithm is studied, and the extracted DCNN features are encoded by Fisher Vectors algorithm, and compared with the locality-constrained linear encoding technique. Experiments show that the feature based on deep convolutional neural network in this paper can obtain higher accuracy than the traditional feature fusion method. According to the result analysis, the tracking accuracy of deep convolutional neural network algorithm is improved under the condition of illumination variation. In the case of local occlusion, the tracking accuracy is also improved.

Physics-informed deep neural network for image denoising.

Image enhancement deep neural networks (DNN) can improve signal to noise ratio or resolution of optically collected visual information. The literature reports a variety of approaches with varying effectiveness. All these algorithms rely on arbitrary data (the pixels' count-rate) normalization, making their performance strngly affected by dataset or user-specific data pre-manipulation. We developed a DNN algorithm capable to enhance images signal-to-noise surpassing previous algorithms. Our model stems from the nature of the photon detection process which is characterized by an inherently Poissonian statistics. Our algorithm is thus driven by distance between probability functions instead than relying on the sole count-rate, producing high performance results especially in high-dynamic-range images. Moreover, it does not require any arbitrary image renormalization other than the transformation of the camera's count-rate into photon-number.

Optimized Deep Convolutional Neural Network for the Prediction of Breast Cancer Recurrence

With more than 2.1 million new cases of diagnosis each year, breast cancer is considered to be the most prevalent women disease. Within 10 years, nearly 30% patients who got cured at early-stages experienced cancer recurrence. Recurrence is a crucial aspect of breast cancer behaviour that is inseparably linked to mortality. Despite its importance, the significant proportion of breast cancer datasets rarely include it, which makes research into its prediction more challenging. It is still difficult to predict who will experience a recurrence and who won't, which has implications for the treatment that goes along with it. Clinicians treating breast cancer may be able to avoid ineffective overtreatment if Artificial Intelligence (AI) methods are developed that can forecast the likelihood of breast cancer recurrence. This work proposes a novel automatic breast cancer recurrence classification and prediction system incorporating novel Deep Convolutional Neural Network (DCNN) algorithm. The proposed DCNN model is deployed on Wisconsin Breast Cancer dataset for further evaluation. The role of AI in forecasting recurrence is examined in this work. The experimental results were analysed for various combination of train and validation dataset. The accuracy, precision, recall and F1-score for the proposed DCNN was calculated as 97.63 %, 98.57 %, 96.84 %, 97.89 % respectively.