Research on the algorithm of urban waste classification and recycling based on deep learning technology

With the development of deep learning, the use of neural network for text detection has been more in-depth research and more widely used. Based on this, this paper studies the Chinese text event detection technology based on improved neural network. In the research, this paper uses the flower pollination algorithm (FPA) to improve the traditional BP neural network algorithm. By optimizing the weights and thresholds of BP neural network, a Chinese text event detection method based on improved neural network is proposed. In order to verify the effect of the Chinese text event detection method based on improved neural network, this paper compares it with the natural scene text detection method, and compares the recall rate, accuracy rate and time-consuming. The results show that the accuracy rate of the natural scene text detection method is 88%, and the recall rate is 73%. The accuracy rate of the text detection method based on the improved neural network is 95% and the recall rate is 86%. The F value of the natural scene text detection method in the Chinese text event detection test is 0.79, which takes 4.56s The value of F in is 0.90, which takes 0.64s. Therefore, the Chinese text event detection method based on improved neural network has better performance.

At present, photovoltaic power generation is becoming more and more important in the field of new energy power generation. The traditional BP neural network algorithm is prone to fall into local extreme points in the training process of the short-term photovoltaic power prediction model, and the training model If the global optimum is not reached, the optimization is stopped, and the weights of the initial network are randomly selected, which leads to problems such as low model prediction accuracy and low iteration efficiency. This paper proposes a short-term photovoltaic power prediction algorithm based on improved BP neural network. The method uses the Levenberg-Marquardt algorithm to replace the traditional gradient descent method for training, which improves the iterative speed of the traditional BP neural network training, and uses the genetic algorithm to optimize the initial weight of the network model, which further improves the model prediction accuracy. The simulation results show that the prediction accuracy of the improved algorithm for short-term photovoltaic power is significantly better than that of the algorithm before the improvement, and the stability and iterative efficiency are greatly improved.

The identification of leaves is of great significance to the identification and classification of plant species and the exploration of the phylogenetic relationship between plants. The accuracy of existing leaf identification technologies needs to be improved. This paper quantifies the image samples, optimizes support vector machine parameters through an improved grid search method, and establishes a GSM-SVM prediction model. The shape features and leaf vein texture are used as discrimination indexes, and the types of leaves are discriminated based on the extracted data information. At the same time, the comparative analysis of traditional BP neural network algorithm is introduced to consider the superiority of the model. Finally, the improved GSM-SVM classification prediction model has a classification accuracy rate of 97.5%, which is obviously better than traditional recognition methods. This method embodies the application of intelligent recognition technology in botany, and it is also conducive to the widespread development of recognition technology.

In this paper, wireless sensors are used to design and build a location management system for smart aging, and the optimization of this system is analysed. The function and structure of the positioning system are designed and implemented. Next, the system commissioning process and field implementation results are summarized and analysed to guarantee the reliability and integrity of the system. Finally, the system requirements and implementation are integrated to realize the basic requirements of the IoT‐based elderly positioning management system and reserve the interface for later expansion. Finally, the system excellence and functional usability tested through unit testing as well as comprehensive testing. In this paper, a theoretical analysis of the prediction algorithm is carried out for the disease prediction module in the health monitoring software of the smart senior care system. To improve the accuracy of the prediction, the traditional BP neural network algorithm is optimized using DS evidence theory, thus fusing multiple sets of prediction results obtained from the BP neural network into a more accurate set of data, and the performance of the algorithm before and after the improvement is compared. The IoT‐based home health management for the elderly starts from the health service demand of the elderly, explains the basic concept of IoT technology and home health management for the elderly, and analyses the feasibility of home health management for the elderly and the advantages of IoT technology in‐home health management for the elderly; through the field research, the IoT‐based home health management platform for the elderly is carried out from three aspects of users, business, and technology. The design covers the platform architecture and functional modules of the IoT‐based senior home health management platform, which can solve the problem that the elderly can spend a comfortable life in their old age without going home.

In high-voltage switchgear, partial discharge (PD) detection using six-degree-of-freedom (6-DOF) manipulators presents challenges. However, these involve inverse kinematics (IK) solution redundancy and the lack of synergistic optimization between end-effector positioning accuracy and energy consumption. To address these issues, a dual-layer adaptive optimization model integrating multiple algorithms is proposed. In the first layer, a spatio-temporal correlation particle memory-based particle swarm optimization BP neural network (STPSO-BP) is employed. It replaces traditional IK, while long short-term memory (LSTM) predicts particle movement trends, and trajectory similarity penalties constrain search trajectories. Thereby, positioning accuracy and adaptability are enhanced. In the second layer, a chaotic mapping-based simulated annealing (CM-SA) algorithm is utilized. Chaotic joint angle constraints, dynamic weight adjustment, and dynamic temperature regulation are incorporated. This approach achieves collaborative optimization of energy consumption and positioning error, utilizing cubic spline interpolation to smooth the joint trajectory. Specifically, the positioning error decreases by 68.9% compared with the traditional BP neural network algorithm. Energy consumption is reduced by 60.18% in contrast to the pre-optimization state. Overall, the model achieves significant optimization. An innovative solution for synergistic accuracy–energy control in 6-DOF manipulators for PD detection is offered.

Traditional image recognition algorithms often face problems such as low recognition accuracy and insufficient robustness when facing complex scenes and multi-class image data. To this end, a hierarchical discriminant analysis (HDA)-based image recognition algorithm was proposed, which effectively improves image recognition performance by constructing a multi-scale feature extraction module, principal component analysis (PCA) dimensionality reduction, attention mechanism, and dynamic hierarchical adjustment strategy combined with a hierarchical feature extraction and discrimination model. The experiment was conducted on three public datasets: CIFAR-10, ImageNet subset (selecting 100 categories with a total of 150000 images, based on covering common object categories and moderate data volume for fair validation of algorithm performance), and MNIST. The performance was compared with models such as VGG16, ResNet50, SVM, KNN, Hierarchical CNN, EfficientNet, GoogLeNet, etc. The results indicated that the proposed method had higher recognition accuracy than other comparative algorithms on different datasets, with accuracies exceeding 90%. The proposed method performed better in terms of mean absolute error and root mean squared error. The F1 value curve of the proposed method was located at the top of the coordinate axis, reaching a maximum value of 92.39%, which was 14.56% higher than the lowest value of 78.24% in the EfficientNet model. This algorithm has better recognition accuracy than traditional algorithms on multiple public datasets, and has strong anti-interference ability and robustness, which can provide reference for optimizing the accuracy of image recognition.

Traditional image object classification and detection algorithms and strategies cannot meet the problem of video image acquisition and processing. Deep learning deliberately simulates the hierarchical structure of human brain, and establishes the mapping from low-level signals to high-level semantics, so as to achieve hierarchical feature representation of data. Deep learning technology has powerful visual information processing ability, which has become the forefront technology and domestic and international research hotspots to deal with this challenge. In order to solve the problem of target space location in video surveillance system, time-consuming and other problems, in this paper, we propose the algorithm based on RNN-LSTM deep learning. At the same time, according to the principle of OpenGL perspective imaging and photogrammetry consistency, we use 3D scene simulation imaging technology, relying on the corresponding relationship between video images and simulation images we locate the target object. In the 3D virtual scene, we set up the virtual camera to simulate the imaging processing of the actual camera, and the pixel coordinates in the video image of the surveillance target are substituted into the simulation image, next, the spatial coordinates of the target are inverted by the inverse process of the virtual imaging. The experimental results show that the detection of target objects has high accuracy, which has an important reference value for outdoor target localization through video surveillance images.

Municipal solid waste (MSW) management currently requires critical attention in ensuring the best principles of socio-economic attributes such as environmental protection, economic sustainability, and mitigation of human health problems. Numerous surveys on the waste management system reveal that approximately 90% of the MSW systems are improperly disposing the wastages in open dumps and landfills. Classifying the wastages into biodegradable and non-biodegradable helps converting them into usable energy and disposing properly. The advancements of effective computational approaches like artificial intelligence and image processing provide wide range of solutions for the present problem identified in MSW management. The computational approaches can be programmed to classify wastes that help to convert them into usable energy. Existing methods of waste classification in MSW remain unresolved due to poor accuracy and higher error rate. This paper presents an experimented effective computer vision–based MSW management solution with the help of the Internet of Things (IoT), and machine learning (ML) techniques namely regression, classification, clustering, and correlation rules for the perception of solid waste images. A ground-up built convolutional neural network (CNN) and CNN by the inception of ResNet V2 models trained through transfer learning for image classification. ResNet V2 supports training large datasets in deep neural networks to achieve improved accuracy and reduced error rate in identity mapping. In addition, batch normalization and mixed hybrid pooling techniques are incorporated in CNN to improve stability and yield state of art performance. The proposed model identifies the type of waste and classifies them as biodegradable or non-biodegradable to collect in respective waste bins precisely. Furthermore, observation of performance metrics, accuracy, and loss ensures the effective functions of the proposed model compared to other existing models. The proposed ResNet-based CNN performs waste classification with 19.08% higher accuracy and 34.97% lower loss than the performance metrics of other existing models.

Backpropgation neural network was trained to predict of combustible fraction heating value of MSW from the physical composition. Waste-to-Energy (WtE) is a viable option for municipal solid waste (MSW) management. The influence of the heating value of municipal solid waste (MSW) is very important on the implementation of WtE systems. As MSW is heterogeneous material, direct heating value measurements are often not feasible. In this study an empirical model was developed to describe the heating value of the combustible fraction of municipal solid waste as a function of its physical composition of MSW using backpropagation neural network. Sampling process was carried out at Jatibarang landfill. The weight of each sorting sample taken from each discharged MSW vehicle load is 100 kg. The MSW physical components were grouped into paper wastes, absorbent hygiene product waste, styrofoam waste, HD plastic waste, plastic waste, rubber waste, textile waste, wood waste, yard wastes, kitchen waste, coco waste, and miscellaneous combustible waste. Network was trained by 24 datasets with 1200, 769, and 210 epochs. The results of this analysis showed that the correlation from the physical composition is better than multiple regression method .

Aiming at the problems of traditional sparse representation algorithms that has a large amount of calculation and low recognition rate under the condition of small samples, a sparse representation classification algorithm (PCA-SRC) combined with principal component feature extraction is proposed. The algorithm can maintain the manifold structure of the finger vein space, solve the singularity problem of the characteristic equation matrix in small sample image recognition. Principal component analysis (PCA) is used to reduce the dimension of the feature matrix to eliminate redundancy. Construct an over-complete dictionary of sparse representation, and design a classifier based on sparse representation to get the classification results. The finger vein feature matching experiment is carried out using the self-built finger vein data set, and the algorithm proposed in this paper is compared with the traditional algorithm. The results show that the recognition rate of the four algorithms can reach more than 94%. The fusion of PCA and SRC speeds up the recognition speed. The recognition rate is higher than the traditional algorithm under different numbers of training samples, and the best recognition rate is 100%. The algorithm has high finger vein recognition rate and low dependence on the training sample set, which provides a new idea for the realization of finger vein recognition.

A fuzzy min-max neural network based on data core (DCFMN) is proposed for pattern classification. A new membership function for classifying the neuron of DCFMN is defined in which the noise, the geometric center of the hyperbox, and the data core are considered. Instead of using the contraction process of the FMNN described by Simpson, a kind of overlapped neuron with new membership function based on the data core is proposed and added to neural network to represent the overlapping area of hyperboxes belonging to different classes. Furthermore, some algorithms of online learning and classification are presented according to the structure of DCFMN. DCFMN has strong robustness and high accuracy in classification taking onto account the effect of data core and noise. The performance of DCFMN is checked by some benchmark datasets and compared with some traditional fuzzy neural networks, such as the fuzzy min-max neural network (FMNN), the general FMNN, and the FMNN with compensatory neuron. Finally the pattern classification of a pipeline is evaluated using DCFMN and other classifiers. All the results indicate that the performance of DCFMN is excellent.

A common way to solve the problem of how to obtain the optimal experimental conditions and the experimental results based on limited experimental data is that use of the traditional genetic algorithm optimization BP neural network algorithm (BPGA). However, the experimental results are often unstable and the error of the optimal results is also relatively large. Therefore, we propose a method of sinusoidal adaptive genetic algorithm (SAGA) double optimization BP neural network. The simulation results show that our proposed algorithm effectively improves the accuracy of the extreme value optimization of the nonlinear function. The network prediction error reaches -0.008 to 0.008, the average extreme value is accurate to 0.0084, and the optimal extreme value is accurate to 0.00007.

In recent years, the research of Multi-Agent Systems (MAS) has become a hot spot in complex systems research. It is widely used in multi-manipulator cooperative equipment, parallel computing, and network resource allocation. The planning of a multi-agent system is more and more critical. This paper takes RoboCup, a classic MAS, as an example. First, it is modeled mathematically through the Markov decision-making process, and a new ball-passing strategy solution based on a deep neural network is proposed. In addition, this paper also tries to use the traditional single-layer neural network to solve the same problem and compares the accuracy difference and model performance of the two types of neural networks. Finally, the possibility of applying a deep learning network in MAS is analyzed. The results show that the deep neural network has high effectiveness in solving the problem of the passing strategy of RoboCup. The ball control rate in the simulation environment is far more than the traditional algorithm, and the model trained by the deep learning network is significantly higher than the traditional neural network in accuracy, efficiency, and other aspects. Applying deep learning networks in other MAS is also likely to have a good effect. In the future, deep learning networks will be an essential planning algorithm.

In recent years, the fall detection system has become an important topic in the homecare system. Compared with the traditional fall detection algorithm, the method used by neural network is more robust and has higher accuracy. However neural network consumes a large amount of energy due to a huge number of computations, and needs more memory to store parameters as compared to traditional algorithms. In this paper, we propose a fall detection system in combination of the traditional algorithm with the neural network. First, we propose a skeleton information extraction algorithm, which transforms depth information into skeleton information and extracts the important joints related to fall activity. Also we have modified the skeleton-based method with seven highlight feature points. Second, we propose a highly robust deep convolution neural network architecture, which uses a pruning method to reduce parameters and calculations in the network. The low number of parameters and calculations makes the system suitable for the implementation on an embedded system. The experiment results show the high accuracy and robustness on the popular benchmark dataset NTU RGB+D. The proposed system has been implemented on NVIDIA Jetson Tx2 platform with real-time processing.

The license plate recognition is an important part of the intelligent traffic management system, and the application of deep learning to the license plate recognition system can effectively improve the speed and accuracy of recognition. Aiming at the problems of traditional license plate recognition algorithms such as the low accuracy, slow speed, and the recognition rate being easily affected by the environment, a Convolutional Neural Network- (CNN-) based license plate recognition algorithm-Fast-LPRNet is proposed. This algorithm uses the nonsegment recognition method, removes the fully connected layer, and reduces the number of parameters. The algorithm—which has strong generalization ability, scalability, and robustness—performs license plate recognition on the FPGA hardware. Increaseing the depth of network on the basis of the Fast-LPRNet structure, the dataset of Chinese City Parking Dataset (CCPD) can be recognized with an accuracy beyond 90%. The experimental results show that the license plate recognition algorithm has high recognition accuracy, strong generalization ability, and good robustness.