Recursive Convolutional Neural Network Research Articles

Deep Learning based Apple Fruit Disease Detection using Dense Net Recursive Convolutional Neural Network (DNRCNN)

Every year, fruit diseases cost the apple industry a lot of money. It can be challenging for growers to identify various apple infections because the symptoms of various illnesses are often similar and may overlap. In this study, we suggest a deep learning-based method for identifying and categorizing apple diseases. Dataset generation, which includes data collection and data labelling, is the first stage of the investigation. On the prepared dataset, we then trained a deep learning-based Dense Net Recurrent Convolutional Neural Network (DNRCNN) model for automatically classifying apple diseases. The end-to-end learning algorithm DNRCNN is appropriate for a range of tasks including image classification, object detection, and segmentation because it automatically extracts complex features from source images and learns them directly. Initialize the parameters of the proposed deep model using transfer learning. To avoid over-fitting, data augmentation techniques like rotation, translation, reflection, and scaling are also used. On the prepared dataset, the proposed Dense Net Recursive Convolutional Neural Network (DNRCNN) model achieves promising results, with an accuracy of about 96%. Some of the intricate and helpful image characteristics for detection are captured by the suggested model for classification. The model can learn the higher-order features of two adjacent layers that are not in the same channel but have a high correlation more effectively than existing techniques. High training and validation accuracy have been achieved when training and validating the suggested model. The findings support the method's usefulness in categorizing different apple diseases and show that it can be useful for farmers

Identification of full-field wind loads on buildings using a mechanism-inspired recursive convolutional neural network with partial structural responses

Indirect identification approaches through structural responses have proven effective for wind load estimation in real-world engineering. Currently, methods for identifying wind loads mainly rely on theoretical inverse identification, with rare research based on the mapping relationship between structural responses and wind loads through machine learning. In this paper, a scheme for identifying full-field wind loads using a recursive convolutional neural network (CNN) inspired by physical mechanisms is proposed. The recursive form of the network, as well as the inspiration for its inputs and outputs, is inspired by the spatial correlation and the mapping relationship between wind loads and structural responses. Thus, the network inputs comprise a fusion of structural acceleration and inter-story displacement responses, while the network outputs represent the independent wind loads on structures. Notably, mismatch test is employed by the network, wherein the training and testing datasets originate from entirely different sources. Specifically, during training, Gaussian white noises that simulate wind loads are utilized, while real wind load data are used for testing. The generalization of the proposed scheme is demonstrated through the identification of full-field wind loads generated by different stationary or non-stationary wind spectra of the 76-story wind-excited benchmark building. Furthermore, the proposed scheme is validated by identifying the full-field wind loads of a 67-story shear wall structure with wind tunnel test data.

Time-Shifted Gramian Angular Field and Recursive Plot Convolutional Neural Network to Adapt Solar Cell Dataset Overfitting in Hybrid Power Generation

Time-Shifted Gramian Angular Field and Recursive Plot Convolutional Neural Network to Adapt Solar Cell Dataset Overfitting in Hybrid Power Generation

Single-image super-resolution based on an improved asymmetric Laplacian pyramid structure

Large-scale factor image super-resolution whose scale factor is greater than 4 is significant in real-world applications for single image super-resolution. The current image super-resolution techniques for large-scale factor, however, frequently upsample low-resolution images in a single pass, leading to edge artifacts in the reconstructed images. In this article, we provide an improved asymmetric Laplacian pyramid network to further realize large-scale factor image super-resolution and fully utilize various size features. A distinct architecture is applied at each layer of the pyramid for improving feature extraction. We extend the first layer of the pyramid with a lightweight transformer design, which enables the model to efficiently collect the contextual information in the sequence by utilizing the multi-headed attention mechanism. Additionally, we combine improved dense skip connections with recursive operations to form a deep dense recursive convolutional neural network that fuses low-level and high-level features while broadening the network's receptive field. The quantitative and qualitative analysis on benchmark dataset's demonstrate that our method provides superior performance in both PSNR and SSIM and is more in line with human visual perception.

Recursive lightweight convolutional neural networks that make noisy images purer and purer

Recursive lightweight convolutional neural networks that make noisy images purer and purer

Effective Vehicle Detection Using Improved Faster Recursive Convolutional Neural Network Model

Effective Vehicle Detection Using Improved Faster Recursive Convolutional Neural Network Model

English Teaching in Artificial Intelligence-based Higher Vocational Education Using Machine Learning Techniques for Students’ Feedback Analysis and Course Selection Recommendation

Higher vocational education is a self-contained method of higher education that is aligned with global productivity and economic development. Its goal is to develop talented workers who contribute significantly to the economy and industry. Teaching analysis, teaching strategy, teaching practice, and assessment are all part of the course design process in high vocational education. Teaching assessment is one of the most effective methods for improving the quality of course teaching among teaching processes. This research proposes novel techniques in English teaching based on artificial intelligence for course selection based on students' feedback. Here, the dataset has been collected based on the students’ feedback on courses for Higher Vocational Education in English teaching. This dataset has been processed to remove invalid data, missing values, and noise. The processed data features have been dimensionality reduction integrated with K-means neural network. And the extracted features have been classified with higher accuracy using recursive elimination-based convolutional neural network. Based on this feedback data classification, recommendation for courses in Higher Vocational Education in English teaching has been suggested. The experimental analysis shows various students' feedback dataset validation and training in terms of accuracy of 96%, precision of 92%, recall of 93%, RMSE of 68%, and computational time of 65%.

Design of deep recursive CNN model for detecting and classifying peston plant

Automatic pest on plant detection in early stage is very essential for food quality control in the agriculture industry. However, the visual method to identify pest on every plant by human is a cumbersome process and cannot be well suited in the agriculture field, because it is time consuming, less accurate and labor intensive. Pest on Plant and plant leaf disease are the major factors responsible for reducing the quality and quantity of food production. Detection at the earlier stage of pest growth and its killing would result in reducing its effect on plant and enhance the quality of food production. Various existing ways have been used to identify and classify pest on plant, but issues have not been resolved, and there is still a scope for improvement. This paper proposes a Deep Recursive Convolutional Neural Networks (DR-CNN) to improve the average running time and achieve high accuracy. DR-CNN model is integrating the convolution, ReLU and Max pooling Layer into single unit and call recursively.

Spam text classification using LSTM Recurrent Neural Network

Sequence Classification is one of the on-demand research projects in the field of Natural Language Processing (NLP). Classifying a set of images or text into an appropriate category or class is a complex task that a lot of Machine Learning (ML) models fail to accomplish accurately and end up under-fitting the given dataset. Some of the ML algorithms used in text classification are KNN, Naïve Bayes, Support Vector Machines, Convolutional Neural Networks (CNNs), Recursive CNNs, Recurrent Neural Networks (RNNs), Long Short Term Memory (LSTM), etc. For this experimental study, LSTM and a few other algorithms were chosen for a more comparative study. The dataset used is the SMS Spam Collection Dataset from Kaggle and 150 more entries were additionally added from different sources. Two possible class labels for the data points are spam and ham. Each entry consists of the class label, a few sentences of text followed by a few useless features that are eliminated. After converting the text to the required format, the models are run and then evaluated using various metrics. In experimental studies, the LSTM gives much better classification accuracy than the other machine learning models. F1-Scores in the high nineties were achieved using LSTM for classifying the text. The other models showed very low F1-Scores and Cosine Similarities indicating that they had underperformed on the dataset. Another interesting observation is that the LSTM had reduced the number of false positives and false negatives than any other model.

Cascaded and Recursive ConvNets (CRCNN): An effective and flexible approach for image denoising

Recently, discriminative learning methods have gained substantial interest in solving inverse imaging problems due to their decent performance and fast inferencing capability. Those methods need separate models for specific noise levels, which in turn require multiple models to be trained to denoise an image. However, images exhibit spatial variant noise which limits the applicability of such methods. In addition, the discriminative learning methods introduce artifacts such as blurring, deblocking, and so forth while denoising an image. To address these issues, we propose a cascaded and recursive convolutional neural network (CRCNN) framework which can cope with spatial variant noise and blur artifacts in a single denoising framework. The CRCNN takes into account down-sampled sub-images for fast inferencing along with the noise level map. We adopt the hybrid orthogonal projection and estimation method on the convolutional layers to improve the generalization capability of the network in terms of non-uniform and spatial variant noise levels. In contrast to the existing methods, the CRCNN framework allows both denoising and deblurring of images using a single framework which preserves the fine details in a denoised image. Extensive experiments have been conducted to validate the effectiveness and flexibility of the CRCNN framework on real as well as synthetic noisy images in comparison to the state-of-the-art denoising methods. The results show that the CRCNN performs effectively on both synthetic as well as spatial variant noise-induced images, thus, proving the practicability of the framework.

RSPCN: Super-Resolution of Digital Elevation Model Based on Recursive Sub-Pixel Convolutional Neural Networks

The digital elevation model (DEM) is known as one kind of the most significant fundamental geographical data models. The theory, method and application of DEM are hot research issues in geography, especially in geomorphology, hydrology, soil and other related fields. In this paper, we improve the efficient sub-pixel convolutional neural networks (ESPCN) and propose recursive sub-pixel convolutional neural networks (RSPCN) to generate higher-resolution DEMs (HRDEMs) from low-resolution DEMs (LRDEMs). Firstly, the structure of RSPCN is described in detail based on recursion theory. This paper explores the effects of different training datasets, with the self-adaptive learning rate Adam algorithm optimizing the model. Furthermore, the adding-“zero” boundary method is introduced into the RSPCN algorithm as a data preprocessing method, which improves the RSPCN method’s accuracy and convergence. Extensive experiments are conducted to train the method till optimality. Finally, comparisons are made with other traditional interpolation methods, such as bicubic, nearest-neighbor and bilinear methods. The results show that our method has obvious improvements in both accuracy and robustness and further illustrate the feasibility of deep learning methods in the DEM data processing area.

Color correction and restoration based on multi-scale recursive network for underwater optical image

Underwater image processing has played an important role in various fields such as submarine terrain scanning, submarine communication cable laying, underwater vehicles, underwater search and rescue. However, there are many difficulties in the process of acquiring underwater images. Specifically, the water body will selectively absorb part of the light when light travels through the water, resulting in color degradation of underwater images. At the same time, due to the influence of floating substances in the water, the light has a certain degree of scattering, which will bring serious problems such as blurred details and low contrast to underwater images. Therefore, using image processing technology to restore the real appearance of underwater images has a high practical value. In order to solve the above problems, we combine the color correction method with the deblurring network to improve the quality of underwater images in this paper. Firstly, aiming at the problem of insufficient number and diversity of underwater image samples, a network combined with depth image reconstruction and underwater image generation is proposed to simulate underwater images based on the style transfer method. Secondly, for the problem of color distortion, we propose a dynamic threshold color correction method based on image global information combined with the loss law of light propagation in water. Finally, in order to solve the problem of image blurring caused by scattering and further improve the overall image clarity, the color-corrected image is reconstructed by a multi-scale recursive convolutional neural network. Experiment results show that we can obtain images closer to underwater style with shorter training time. Compared with several latest underwater image processing methods, the proposed method has obvious advantages in multiple underwater scenes. Simultaneously, we can restore the color information, remove blurring and boost detail for underwater images.

Recursive convolutional neural networks in a multiple-point statistics framework

This work proposes a new technique for multiple-point statistics simulation based on a recursive convolutional neural network approach coined RCNN. The work focuses on methodology and implementation rather than performance to demonstrate the potential of deep learning techniques in geosciences. Two and three dimensional case studies are carried out. A sensitivity analysis is presented over the main RCNN structural parameters using a well-known training image of channel structures in two dimensions. The optimum parameters found are applied into image reconstruction problems using two other training images. A three dimensional case is shown using a synthetic lithological surface-based model. The quality of realizations is measured by statistical, spatial and accuracy metrics. The RCNN method is compared to standard MPS techniques and an improving framework is proposed by using the RCNNE-type as secondary information. Strengths and weaknesses of the methodology are discussed by reviewing the theoretical and practical aspects.

SAR Image Despeckling Employing a Recursive Deep CNN Prior

Synthetic aperture radar (SAR) images are inherently affected by speckle noise, for which deep learning-based methods have shown good potential. However, the deep learning-based methods proposed until now directly map low-quality images to high-quality images, and they are unable to characterize the priors for all the kinds of speckle images. The variational method is a classic model optimization approach that establishes the relationship between the clean and noisy images from the perspective of a probability distribution. Therefore, in this article, we propose the recursive deep convolutional neural network (CNN) prior model for SAR image despeckling (SAR-RDCP). First, the data-fitting term and regularization term of the SAR variational model are decoupled into two subproblems, i.e., a data-fitting block and a deep CNN prior block. The gradient descent algorithm is then used to solve the data-fitting block, and a predenoising residual channel attention network based on dilated convolution is used for the deep CNN prior block, which combines an end-to-end iterative optimization training. In the experiments undertaken in this study, the proposed model was compared with several state-of-the-art despeckling methods, obtaining better results in both the quantitative and qualitative evaluations.

Recursive multiresolution convolutional neural networks for 3D aortic valve annulus planimetry.

Transcatheter aortic valve replacement (TAVR) is the standard of care in a large population of patients with severe symptomatic aortic valve stenosis. The sizing of TAVR devices is done from ECG-gated CT angiographic image volumes. The most crucial step of the analysis is the determination of the aortic valve annular plane. In this paper, we present a fully tridimensional recursive multiresolution convolutional neural network (CNN) to infer the location and orientation of the aortic valve annular plane. We manually labeled 1007 ECG-gated CT volumes from 94 patients with severe degenerative aortic valve stenosis. The algorithm was implemented and trained using the TensorFlow framework (Google LLC, USA). We performed K-fold cross-validation with K = 9 groups such that CT volumes from a given patient are assigned to only one group. We achieved an average out-of-plane localization error of (0.7 ± 0.6) mm for the training dataset and of (0.9 ± 0.8) mm for the evaluation dataset, which is on par with other published methods and clinically insignificant. The angular orientation error was (3.9 ± 2.3)° for the training dataset and (6.4 ± 4.0)° for the evaluation dataset. For the evaluation dataset, 84.6% of evaluation image volumes had a better than 10° angular error, which is similar to expert-level accuracy. When measured in the inferred annular plane, the relative measurement error was (4.73±5.32)% for the annular area and (2.46±2.94)% for the annular perimeter. The proposed algorithm is the first application of CNN to aortic valve planimetry and achieves an accuracy on par with proposed automated methods for localization and approaches an expert-level accuracy for orientation. The method relies on no heuristic specific to the aortic valve and may be generalizable to other anatomical features.

The Image Game: Exploit Kit Detection Based on Recursive Convolutional Neural Networks

Malware has been installed through drive-by downloads via exploit kit attacks. However, the prior signature- or dynamic-based detection approach to the continuously increasing number of suspicious samples is time-consuming. In such circumstances, convolutional neural networks (ConvNets) can help in rapid detection owing to their direct image-feature generation using exploit codes. However, the general ConvNet model entails the vanishing gradient problem, where the features used for a deep learning-based detection method will become less effective as the network is deepened to improve detection accuracy. In this paper, we propose a multiclass ConvNet model to classify exploit kits, where we adopt various image processing techniques and adjust the size and other parameters of images. The proposed ConvNet model recursively updates images and is designed for fully preserving image properties. This model updates the output of feature maps and pooling using an original image. This model was tested using 36,863 real-world datasets, achieving a 98.2% accuracy in exploit kit detection and family classification. Most importantly, the proposed model is 38 times faster than previous machine learning models, and training time is reduced by 77.8% when compared with prior well-known ConvNet models.

Feedback Based Adaptive Recurrent Neural Network for Cancer Detection using Gene Data Pattern

Cancer detecting technology plays a vital role in the medical community. Researches have shown that patients that are affected by cancer carry same type of genetic patterns in their DNA. With this in mind, this research work concentrates on analysing gene pattern for detecting cancer using deep learning algorithms. The Feedback based Adaptive Recurrent Neural Network (FA-RNN) approach is designed to classify and analyse the gene pattern recognition. The data augmentation is done to improve the quality of the input data from COSMIC dataset which includes the detection of missing values, removing the noise present in input using multiple imputations and reducing higher base value can be done using dimensionality reduction process. After obtaining the improved dataset, the training phase begins by estimating the exact weight value of feedback layer using feedback weight loop calculation technique to lessen number of repetition during training. Moreover, the error calculation is done to evaluate the exact weight values of feedback layer used for classification. Finally the classification is done by selecting the next appropriate hidden neuron using the neuron selection activation function. The performance of the Feedback based Adaptive Recurrent Neural Network technique can be analysed using the evaluation metrics accuracy, computation time and Root Mean Square Error (RMSE) and the attained results are compared with the Recursive Neural Network(RNN) and Convolutional Neural Network(CNN) algorithms. The obtained results such as higher accuracy, reduced RMSE and less computation time in Feedback based Adaptive Recurrent Neural Network indicates that it performs the enhanced operation than CNN and RNN.

SCRSR: An efficient recursive convolutional neural network for fast and accurate image super-resolution

Convolutional neural networks have recently demonstrated high-quality reconstruction for single image super-resolution (SR). These CNN networks effectively recover a high-resolution (HR) image from a low-resolution (LR) image, at the cost of enormous parameters and heavy computational burden. In this work, we propose a recursive efficient deep convolutional network for fast and accurate single-image SR with only 0.28M parameters. A Split-Concatenate-Residual (SCR) block is proposed to reduce computation and parameters. With downsampling block and upsampling block, we significantly reduce computational complexity and enlarge the size of the receptive field. Specifically, two-level recursive learning is proposed which can improve accuracy by increasing depth without adding any weight parameters. We also employ local, semi-global and global residual techniques to train our very deep network steadily and improve its performance. Extensive experiments indicate that our proposed method Split-Concatenate-Residual Super Resolution (SCRSR) yields promising SR performance while maintaining shorter running time and fewer parameters.

Transformer-Based Neural Network for Answer Selection in Question Answering

Answer selection is a crucial subtask in the question answering (QA) system. Conventional avenues for this task mainly concentrate on developing linguistic tools that are limited in both performance and practicability. Answer selection approaches based on deep learning have been well investigated with the tremendous success of deep learning in natural language processing. However, the traditional neural networks employed in existing answer selection models, i.e., recursive neural network or convolutional neural network, typically suffer from obtaining the global text information due to their operating mechanisms. The recent Transformer neural network is considered to be good at extracting the global information by employing only self-attention mechanism. Thus, in this paper, we design a Transformer-based neural network for answer selection, where we deploy a bidirectional long short-term memory (BiLSTM) behind the Transformer to acquire both global information and sequential features in the question or answer sentence. Different from the original Transformer, our Transformer-based network focuses on sentence embedding rather than the seq2seq task. In addition, we employ a BiLSTM rather than utilizing the position encoding to incorporate sequential features as the universal Transformer does. Furthermore, we apply three aggregated strategies to generate sentence embeddings for question and answer, i.e., the weighted mean pooling, the max pooling, and the attentive pooling, leading to three corresponding Transformer-based models, i.e., QA-TF $_{{W\!P}}$ , QA-TF $_{{M\!P}}$ , and QA-TF $_{{A\!P}}$ , respectively. Finally, we evaluate our proposals on a popular QA dataset WikiQA. The experimental results demonstrate that our proposed Transformer-based answer selection models can produce a better performance compared with several competitive baselines. In detail, our best model outperforms the state-of-the-art baseline by up to 2.37%, 2.83%, and 3.79% in terms of MAP, MRR, and accuracy, respectively.

Blind Deblurring via a Novel Recursive Deep CNN Improved by Wavelet Transform

Blind image deconvolution is an ill-posed problem, which is mainly addressed by the regularization methods. Wavelet transform is an effective denoising method related to regularized inversion. In this paper, wavelet transform is utilized to decompose and extract the low- and high-frequency information of the blurred image, which is taken as the first step of the presented deblurring methods in this paper. However, when the process highlights the approximate portion of the image feature, the blurry image will be smoothed so that the image is distorted. Simultaneously, the overall wavelet transform will result in excessive data redundancy for the image. Thus, in the second step, based upon the recursive convolution neural network, a deep recursive convolutional neural network (R-DbCNN) is designed, which can eliminate or weaken the characteristics of high data redundancy and image smoothness caused by wavelet transform to remove the blur of corrupted image. Comparing with the traditional convolution neural network in image restoration, R-DbCNN has better performance in deblurring with fast training speed. Thereafter, a novel loss function is built to attain the best deblurring effect of the proposed method based on the L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> regularization term. The experiment results demonstrate that our method has practical applicability for image restoration with different blurs, such as the motion blur, Gaussian blur, and out-of-focus by camera.