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

Waste recycling is a critical issue for environment pollution management while garbage classification determines the recycling efficiency. In order to reduce labor costs and increase garbage classification capacity, a machine vision system is established based on the deep learning and transfer learning. In this new method, an improved MobileNetV2 deep learning model is proposed for garbage detection and classification, where the attention mechanism is introduced into the first and last convolution layers of the MobileNetV2 model to improve the recognition accuracy and the transfer learning uses a set of pre-trained weight parameters to extend the model generalization ability. In addition, the principal component analysis (PCA) is employed to reduce the dimension of the last fully connected layer to enable real-time operation of the developed model on an edge device. The experimental results demonstrate that the proposed method generates 90.7 % of the garbage classification accuracy on the "Huawei Cloud" datasets, the average inference time is 600 ms on the raspberry Pi 4B microprocessor, and the model volume compression is 30.1 % of the basic MobileNetV2 model. Furthermore, a garbage sorting porotype is designed and manufactured to evaluate the performance of the proposed MobileNetV2 model on the real-world garbage identification, which turns out that the average garbage classification accuracy is 89.26 %. Hence, the developed garbage sorting porotype can be used a effective tool for sustainable waste recycling.

A New Kinect V2-Based Method for Visual Recognition and Grasping of a Yarn-Bobbin-Handling Robot.

This work proposes a Kinect V2-based visual method to solve the human dependence on the yarn bobbin robot in the grabbing operation. In this new method, a Kinect V2 camera is used to produce three-dimensional (3D) yarn-bobbin point cloud data for the robot in a work scenario. After removing the noise point cloud through a proper filtering process, the M-estimator sample consensus (MSAC) algorithm is employed to find the fitting plane of the 3D cloud data; then, the principal component analysis (PCA) is adopted to roughly register the template point cloud and the yarn-bobbin point cloud to define the initial position of the yarn bobbin. Lastly, the iterative closest point (ICP) algorithm is used to achieve precise registration of the 3D cloud data to determine the precise pose of the yarn bobbin. To evaluate the performance of the proposed method, an experimental platform is developed to validate the grabbing operation of the yarn bobbin robot in different scenarios. The analysis results show that the average working time of the robot system is within 10 s, and the grasping success rate is above 80%, which meets the industrial production requirements.

Measuring Shape Parameters of Pearls in Batches Using Machine Vision: A Case Study.

To solve the problem of low precision of pearl shape parameters' measurement caused by the mutual contact of batches of pearls and the error of shape sorting, a method of contacting pearls' segmentation based on the pit detection was proposed. Multiple pearl images were obtained by backlit imaging, the quality of the pearl images was improved through appropriate preprocessing, and the contacted pearl area was extracted by calculating the area ratio of the connected domains. Then, the contour feature of the contact area was obtained by edge tracking to establish the mathematical model of the angles between the edge contour points. By judging the angle with a threshold of 60° as the candidate concave point, a concave point matching algorithm was introduced to get the true concave point, and the Euclidean distance was adopted as a metric function to achieve the segmentation of the tangent pearls. The pearl shape parameters' model was established through the pearl contour image information, and the shape classification standard was constructed according to the national standard. Experimental results showed that the proposed method produced a better segmentation performance than the popular watershed algorithm and morphological algorithm. The segmentation accuracy was above 95%, the average loss rate was within 4%, and the sorting accuracy based on the shape information was 94%.

Research on Multi-Dimensional Optimal Location Selection of Maintenance Station Based on Big Data of Vehicle Trajectory.

In order to rationally lay out the location of automobile maintenance service stations, a method of location selection of maintenance service stations based on vehicle trajectory big data is proposed. Taking the vehicle trajectory data as the demand points, the demand points are divided according to the region by using the idea of zoning, and the location of the second-level maintenance station is selected for each region. The second-level maintenance stations selected in the whole country are set as the demand points of the first-level maintenance stations. Considering the objectives of the two dimensions of cost and service level, the location model of the first-level maintenance stations under two-dimensional programming is established, and the improved particle swarm optimization algorithm and immune algorithm, respectively, are used to solve the problem. In this way, the first-level maintenance stations in each region are obtained. The example verification shows that the location selection results for the maintenance stations using the vehicle trajectory big data are reasonable and closer to the actual needs.