An Optimized Faster R-CNN Framework with Golden Eagle Optimization for Garbage Detection and Classification

Garbage classification is a social issue related to people’s livelihood and sustainable development, so letting service robots autonomously perform intelligent garbage classification has important research significance. Aiming at the problems of complex systems with data source and cloud service center data transmission delay and untimely response, at the same time, in order to realize the perception, storage, and analysis of massive multisource heterogeneous data, a garbage detection and classification method based on visual scene understanding is proposed. This method uses knowledge graphs to store and model items in the scene in the form of images, videos, texts, and other multimodal forms. The ESA attention mechanism is added to the backbone network part of the YOLOv5 network, aiming to improve the feature extraction ability of the network, combining with the built multimodal knowledge graph to form the YOLOv5‐Attention‐KG model, and deploying it to the service robot to perform real‐time perception on the items in the scene. Finally, collaborative training is carried out on the cloud server side and deployed to the edge device side to reason and analyze the data in real time. The test results show that, compared with the original YOLOv5 model, the detection and classification accuracy of the proposed model is higher, and the real‐time performance can also meet the actual use requirements. The model proposed in this paper can realize the intelligent decision‐making of garbage classification for big data in the scene in a complex system and has certain conditions for promotion and landing.

Aiming at the problems of unclear, difficult, and inefficient classification of traditional manual waste, and the difficulty of deploying large existing garbage classification network models, a lightweight garbage detection and classification network S-YOLOv5 is designed based on YOLOv5s. First, a garbage dataset containing 18 types of common household garbage is constructed and labeled according to the principles of garbage classification; secondly, a module combining shufflenetv2 and CoordAttention was introduced to replace the YOLOv5s backbone network, and the ReLU activation function in the shufflenet module was substituted by FReLU; finally the PANet structure was replaced by the BiFPN structure, so as to reduce the model complexity and achieve lightweight while maintaining a high mAP. The experimental results show that the size of S-YOLOv5 is only 2.6MB, which is about 1/6 of the original network size, and the mAP is 80.2%. The size of the proposed network is reduced while maintaining high accuracy, making it more suitable for deployment in smart devices.

Garbage detection and classification using a new deep learning-based machine vision system as a tool for sustainable waste recycling

With the improvement of people’s living standards, garbage classification is gradually forced. However, due to people’s awareness and knowledge, the classification accuracy and disposal of garbage are difficult to keep pace with guideline changes. With the consideration of the problems of low efficiency, heavy task and poor environment of garbage manual classification, an improved YOLOv7 target detection method is proposed to realize the effective classification of garbage. In this study, the recursive gated convolutional gnconv was used to establish the HorNet network architecture, and the model was trained by making specific data sets. The C3HB module is added to the YOLO model, and the pooling layer is optimized to replace SPPFCSPC to improve the detection accuracy of the target. The experimental results show that the garbage detection and classification method proposed in this study has excellent accuracy. Experiments show that the map value, accuracy and recall rate of the proposed model on garbage datasets are 99.25%, 99.33% and 98.03%, respectively, which are 1.50%, 3.99% and 1.41% higher than those of YOLOv7. The overall results are better than the original model.

Garbage classification is an essential work in daily life. With the development of artificial intelligence (AI), we have begun to use object detection to achieve garbage classification. However, the stacking and occlusion of garbage often make the detection performance unsatisfactory. In this paper, YOLOv5 algorithm is used for garbage detection and classification. YOLOv5 has three detection heads, which can detect objects at multiple scales. At the same time, Mosaic and Coarse Dropout image augmentation techniques are used to process our private data set of garbage, which improves the diversity of images. In this way, the target classification accuracy of the detection model can reach 75.7%, which is significantly improved compared with 69.9% without data augmentation. This paper provides a new technical support for the scene classification of stacked garbage based on deep learning.

With the rise of deep learning technology, intelligent garbage detection provides a new idea for garbage classification management. However, due to the interference of complex environments, coupled with the influence of the irregular features of garbage, garbage detection in complex scenarios still faces significant challenges. Moreover, some of the existing research suffer from shortcomings in either their precision or real-time performance, particularly when applied to complex garbage detection scenarios. Therefore, this paper proposes a model based on YOLOv8, namely HGCS-Det, for detecting garbage in complex scenarios. The HGCS-Det model is designed as follows: Firstly, the normalization attention module is introduced to calibrate the model's attention to targets and to suppress the environmental noise interference information. Additionally, to weigh the attention-feature contributions, an Attention Feature Fusion module is employed to complement the attention weights of each channel. Subsequently, an Instance Boundary Reinforcement module is established to capture the fine-grained features of garbage by combining strong gradient information with semantic information. Finally, the Slide Loss function is applied to dynamically weight hard samples arising from the complex detection environments to improve the recognition accuracy of hard samples. With only a slight increase in parameters (3.02M), HGCS-Det achieves a 93.6% mean average precision (mAP) and 86 FPS on the public HGI30 dataset, which is a 3.33% higher mAP value than from YOLOv12, and outperforms the state-of-the-art (SOTA) methods in both efficiency and applicability. Notably, HGCS-Det maintains a lightweight architecture while enhancing the detection accuracy, enabling real-time performance even in resource-constrained environments. These characteristics significantly improve its practical applicability, making the model well suited for deployment in embedded devices and real-world garbage classification systems. This method can serve as a valuable technical reference for the engineering application of garbage classification.

Garbage is the result left over from the process of daily human activities and activities which are considered no longer suitable for use, ranging from household waste to large-scale industrial waste. Therefore, the classification of waste is important because the problem of waste disposal is increasing and the way of processing is wrong. This research focuses on the classification of organic and non-organic waste using the DenseNet architecture. The dataset is processed first and each image in the dataset is resized to 128x128 pixels before being used in the model. We then trained all DenseNet types namely DenseNet121, DenseNet169, DenseNet 201, and compared their performance. Based on the test results, all DenseNet models that were trained were able to produce good accuracy, precision, recall, and F1 scores in garbage classification. In particular, our designed DenseNet121 model achieves 93.1 accuracy, 94.08% precision, 94.00% recall, 94.03% F1 score and 1min 34s training time as the best among other models. These results prove that the DenseNet architecture can be used to classify organic and non-organic waste correctly.

In 2017, the Chinese State Council issued the Implementation Plan for Domestic Waste Classification System, mandating waste classification in 46 pilot areas. This led to further development of garbage classification management in Chinese universities. However, most research on garbage classification management in universities has focused solely on education. Using the Smith model for analysis from a public policy perspective, this study identifies four problems with garbage classification management in Chinese universities at all levels: (1) idealistic policies require a long period of practice and a perfect system to be implemented in universities, (2) the old system of implementing agencies hinders the implementation of new policies, (3) the government and universities do not prioritize follow-up policy implementation environments, and (4) there are significant differences in the recognition of the target group. This study proposes three suggestions for the government, society, and schools to address these problems. The government should increase research and development of related public infrastructure, society should unite various social actors to jointly shape a good environment for garbage classification, and schools should guide students to take collective responsibility for garbage classification. Thus, this study proposes a new perspective for the research on garbage classification management in Chinese universities from the perspective of garbage classification policy.

Object detection is an important research task in computer vision. With environment problems becoming seriously, garbage detection has become a hot direction in object detection. Though great progress has been made in garbage detection, there still exists challenges for general-purpose detectors with no public datasets and low recognition accuracy rate. Based on YOLOV4 model, this paper proposes an improved YOLOV4 detector for garbage detection. Specifically, CBAM is added to the feature extraction network in order to better extract features in deep networks, and focal function is integrated into the loss function to improve class imbalance. When preparing the data, environment information and random information are added to pictures to simulate the distribution of garbage in real environment, and finally an urban household garbage dataset containing 47 classes of 45,910 images named TrashSet has been produced. Experiments results on TrashSet verify that our detector has a considerable performance, and mAP reaches 97.15%.

<p indent="0mm">The increasing surge in solid waste generation poses tremendous risks to the safety of the environment, resources, and industry, prompting solid waste management to be a global environmental issue and a grand challenge in the 21st century. Solid waste has a double attribute of resource and environment, which is commonly resolved by recycling and detoxication. Previously, the resource attribute of solid waste was measured with recyclability. To achieve the sound management of solid waste, its environmental attribute discovery plays a crucial role in process selecting and policy formulating, especially for some hazardous industrial waste. In this study, we first proposed the removability concept to indicate the environmental attribute, and then established an environmental attribute measuring method, utilizing an interdisciplinary philosophy of statistical entropy and pollution grade. By measuring the removability of toxic substances towards detoxication, the method determines the potential and difficulty of toxic substance removal and resource recovery under current process conditions. Applying the method, we evaluated the removability of various toxic substances in 24 types of industrial waste. A sensitivity analysis was also picked up to identify the key influencing factors. Finally, we analyzed the feasibility of removing toxic substances from industrial waste by recovery technologies. The research on heavy metals and inorganic non-metallic toxic substances in industrial waste shows that the removability of these toxic substances is correlated with valence and stability. The results range from 11 bit^–1 to 66 bit^–1, revealing a low removability of the above toxic substances. However, the removability of organic toxic substances is related to their degradability. Besides, the combination of heavy metals and organic toxic substances could further decrease the removability of the toxic substances. The sensitivity analysis assessed the influence of different factors on the removability of toxic substances. The results suggest that the component, stability, and concentration of toxic substances have obvious effects on removability. To be specific, lower stability will reduce the difficulty to enrich and remove toxic substances. Thus, the high stability of the toxic substances in industrial waste emphasizes the requirement of an additional energy supply to change the phase composition or valence state of the toxic substances. However, the improvement of removability has an efficiency limit, with the critical value of the limit relating to the initial removability. Compared to virgin mining, the removability of over 50% of toxic substances in industrial waste fails to meet the requirements under the condition of the lowest industrial grade. Meanwhile, the feasibility of toxic substance removal and resource recovery in industrial waste detoxication should take both the types of waste and toxic substances into account. From the perspective of waste, only 6 types of industrial waste among 24 types of industrial waste being studied have a removal index greater than zero, including gold tailings, steel slag, and fly ash. Hence, it is suitable to enable toxic substance removal and impair the environmental attribute by material recovery when treating these 6 types of waste. In contrast, the rest types of industrial waste are of high environmental attributes, making it hard to remove the toxic substances by recovery due to the inferior removability and economic feasibility under the current technological condition. From the perspective of toxic substances, the slight differences of toxic substances in industrial wastes make it difficult to perform targeted separation and precise removal with the same technology. In a conclusion, removability is an intrinsic property of solid waste that can influence extrinsic properties, such as the environmental risk of solid waste sound management. Therefore, we considered incorporating the environmental attribute into the classification of solid waste and carried out the complete process of detoxification and accurate accounting. It will assist the recycling industry in the design of waste treatment and resource recycling, and guide the upgrading of waste disposal technology.

The overwhelming amount of household solid waste generated daily is alarming, and this contributes to the rise in pollution and drastic climate change. In such a context, automated waste classification at the initial stage of disposal can be an effective solution to separate recyclable items. Convolutional Neural Networks based on deep learning are often used for automated waste classification, but however, research works are limited to insufficient categories of waste such as the TrashNet dataset consisting of 2,527 images and 6 categories of waste. This dataset does not include other important categories such as battery, biological, and clothing items to reflect real-life environmental problems. Therefore, in this paper, a larger dataset consisting of 15,515 images and 12 categories of common household solid waste was used to evaluate the performance of DenseNet121, DenseNet169, EfficientNetB0, InceptionV3, MobileNetV2, ResNet50, VGG16, VGG19, and Xception Convolutional Neural Network models. Data augmentation was applied to solve the problem of class imbalance, and findings of my first research showed that the Xception model compiled with Adam optimiser outperformed all other models with an accuracy of 88.77% and an F1-score of 0.89. The performance of the model was improved to 89.57% with an F1-score of 0.90 when compiled with Nadam optimiser. However, further experimentation showed that the model did not generalise well despite reaching an accuracy of 93.42% and an F1-score of 0.93 when trained without data augmentation. This demonstrates the feasibility of the proposed model for real-life environmental problems.

This paper aims to solve the problems of low operation efficiency of the traditional urban garbage sorting and recycling logistics network and worrying about the sanitation environment based on social construction. It will be connected to the front end, and the model innovation will be carried out with the collection and transportation logistics of the key link of garbage classification in the end recycling process. The method adopted is to establish a complete system, using differential centrifugal and pneumatic power transportation methods to act on garbage classification and pipeline transportation, formulate standardized collection and transportation procedures, and implement block-based independent management on a community basis. Based on the above methods, the one-stop service of residents' domestic waste collection, classification, recycling, compression, packaging, and transportation is completed, solving the problem of urban waste and finally forming the result of constructing an intelligent process of garbage collection, transportation and classification. This paper uses the domestic waste rail logistics system innovation to provide smarter processing ideas for future communities and create a better living environment for residents. It has a positive effect on implementing urban organic construction, promoting industrial upgrading and economic development.

Today’s garbage classification is not fully automated and mostly relies on manual labor. In order to efficiently classify garbage and increase resource re-usability, society needs a modern solution. The development of artificial intelligence, especially in object detection, provides an excellent opportunity to develop real-time accurate garbage detection. This project is about real-time garbage detection using YOLOX and is used to detect seventeen categories of garbage. After the camera detects the garbage, it will automatically identify and analyze the type and specific the position of the garbage, which will be automatically picked up by a mechanical jaws. The datasets used in this paper consists of 5000 images with label. With the officially supported TensorRT optimization and acceleration technology, the model can be used in low-cost embedded devices and can detect garbage in near real-time without the need for a stable Internet connection. Our evaluation and comparison of these models includes some key metrics such as mean accuracy (mAP), inference time per frame, floating point operations(FLOPs), and the number of parameters of the model. The test results show that YOLOX’s mAP <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.5:0.95</sub> can exceed 97% and the inference speed can exceed 32 fps. The performance far exceeds that of other YOLO series, with high accuracy and speed in complex environments, making it highly applicable to today’s garbage classification needs.

With the increasing demand for garbage classification, images and videos are required to complement each other in real-time scenarios. Therefore, the research of garbage detection and classification is still of long-term value. This paper introduces a novel garbage detection and sorting framework called MCMM that incorporates multi-column convolution and matrix-multiplication (MatMul)-free based Transformer. Specially, the multi-column convolution is designed to enhance the performance of image processing tasks through multi-scale feature extraction and adaptation to objects of varying sizes. MatMul-free Transformer significantly reduces computational complexity and hardware overhead by eliminating matrix multiplication, while maintaining high model performance. The experimental evaluation shows that the MCMM network achieves better results in both qualitative and quantitative methods compared to existing competing methods.

Domestic waste classification is of great significance for building a beautiful living environment. Waste classification is a scientific management method for effective disposal of waste. Aiming at the problem of garbage pollution faced by human beings at present, a garbage classification algorithm based on deep learning is proposed.In this paper, we first build a garbage classification dataset, identify the categories of garbage items in the picture, and then build a garbage classification model. Through the comparison of different models, we test the accuracy of different models on the garbage classification results, and find a more effective and accurate garbage classification algorithm model.The experiment shows that this algorithm analyzes the existing model, preprocesses the data image, and then uses the image enhancement method to increase the robustness of the model used. It improves the model by adjusting the learning rate, label smoothing, and data normalization processing strategies. The optimized algorithm precision increases to 95%, and makes the model have better generalization ability.