Single-stage point cloud 3-dimensional object detection algorithms based on pillars have gained significant attention and widespread application in the industry due to their high operational efficiency. However, the loss of fine-grained information loss in 3-dimensional features of point clouds caused by pillar-based quantization results in weaker detection capabilities for small objects in sparse point clouds. Although some studies have proposed solutions to this problem, they often come at the cost of either greater detection time or compromised detection accuracy for large targets. For this reason, this paper proposed an enhanced pillar-based point cloud object detection algorithm with enhanced pillar feature encoding. Firstly, a pillar feature encoding network is constructed to combine local and global features of point clouds within pillar cells, enhancing the representation capability of pillar-quantized features. Then, a backbone network that combines 2-dimensional sparse convolutional blocks with a feature fusion network was designed to fuse multi-scale high-level abstract semantic features and low-level fine-grained spatial features, preventing excessive focus on small-size features and thus degrading the detection performance for large targets. Lastly, the model was trained and tested on the KITTI autonomous driving dataset, with experimental results visualized and ablation studies conducted. The results show that, the proposed algorithm, under the medium difficulty level of the KITTI dataset, has an average precision mean of 63.54% across multiple categories, an average orientation similarity mean of 70.72%, and an average detection frame rate of 31.5 f/s. Compared with the PointPillars, TANet, and PiFEnet, the average precision mean of the algorithm proposed in this paper has increased by 2.44, 2.05, and 2.38 percentage points respectively, and the average orientation similarity mean has increased by 4.69, 0.68, and 7.83 percentage points respectively, demonstrating potential for engineering applications in comparisons with similar algorithms.

Open World Object Detection (OWOD) extends the problem of object detection to more complex real-world dynamic scenarios, requiring the system to recognize all known and unknown object categories in the image and possess the capability for incremental learning based on newly introduced knowledge. However, current OWOD methods typically mark regions with high object scores as unknown objects and largely rely on supervision of known objects. Although these methods can detect unknown objects that are similar to known ones, they suffer from a significant label bias problem, where regions dissimilar to known objects are often misclassified as part of the background. To address this issue, this study first proposed an unsupervised region proposal generation method based on a large visual model to enhance the model’s ability to detect unknown objects. Then, considering that the sensitivity of the Region of Interest (ROI) classification stage to new categories during model training can affect the generalization performance of the Region Proposal Network (RPN) in the proposal generation stage, a decoupled joint training method for RPN region proposal generation and ROI classification was introduced to improve the model's capability to resolve label bias problems. Experimental results show that the method proposed in this study has achieved a significant improvement in detecting unknown objects on the MS-COCO dataset, with the unknown category recall rate exceeding that of the previous SOTA methods by more than twice, reaching 52.1%, while maintaining competitiveness in detecting known object categories. In terms of inference speed, the model, constructed using pure convolutional neural networks rather than dense attention mechanisms, achieves a frame rate 8.18 f/s higher than that of deformable DETR-based methods.

Fast Fourier transform (FFT) algorithm finds widespread application in scientific computing and related fields. To fully leverage the computational power of the GPU and further enhance the performance of FFT calculations, this paper proposed a high-performance multi-dimensional FFT computation scheme based on the Matrix Core for the matrix form of Stockham FFT. In terms of computational optimization, this scheme utilizes Matrix Core to accelerate matrix multiplications in FFT computation while leveraging compiler intrinsic instructions to perform small-grained matrix multiply-accumulate operations, enabling Matrix Core to support FFT computations of more sizes. To minimize memory access, the proposed scheme directly performs matrix element-wise multiplication operations in the registers according to the distribution pattern of Matrix Core’s data across thread registers. It also mitigates bank conflicts by reordering data in shared memory, adopts a double-buffering strategy to alleviate access bottlenecks, and proposes an efficient matrix transposition strategy to accelerate multidimensional FFT computations. In this paper, the proposed scheme was compared to the well-known high-performance FFT computation libraries rocFFT and VkFFT on the AMD MI250 GPU platform. The results demonstrate that the proposed scheme outperforms rocFFT and VkFFT in terms of average computational performance for 1-dimensional, 2-dimensional, and 3-dimensional FFTs on the AMD MI250 GPU platform. For 3D FFT calculation, this method has an average performance that is 1.5 times faster than rocFFT and 2.0 times faster than VkFFT, demonstrating significant performance improvements.

Image inpainting refers to the process of filling in missing regions of an image with plausible content, which is one of the significant issues in the fields of computer vision and image processing research. Current research on image inpainting algorithms has made substantial progress. However, when dealing with complex images with large missing areas, existing algorithms still face challenges in generating high-quality complete images due to the lack of effective network structures to capture long-range dependencies and high-level semantic information in the images. To address the issue of large-scale missing image inpainting, this paper proposed an image inpainting algorithm based on hybrid encoding and mask spatial modulation. The aim is to expand the limited receptive field of image inpainting networks, effectively obtain global information from the visible regions of the image, and fully utilize the effective information from the visible regions. Firstly, a hybrid encoding network was used to extract local and global information features from the visible regions of the image. Then, a mask spatial modulation module dynamically adjusted the diversity in generating missing regions based on the size of the missing area. Finally, a method based on StyleGAN2 was used to generate complete images. Experimental results show that the proposed algorithm can effectively handle images with large-scale missing areas, generating high-quality images with diversity, and can be applied to data augmentation in visual saliency models.

With the deepening application of artificial intelligence technology in the field of customer service, telecommunications operators have raised higher standards for the accuracy of AI service knowledge recommendations. To enhance the efficiency and accuracy of knowledge recommendation in telecommunications operators’AI customer service systems, this paper proposed a large-scale customer service knowledge recommendation model driven by intent understanding. Firstly, the synonym and dialogue sequence keyword extraction model was employed to identify key terms in user queries. These keywords were then matched with questions in a standard question bank using semantic similarity comparison techniques to generate the most relevant standard questions. Additionally, a generative agent technology framework was utilized to construct and enrich the standard question bank, enabling the automatic generation of knowledge questions. The extracted standard questions were input into the ChatGLM2-6B large language model, which has been pre-trained and aligned with human preferences, further improving the accuracy of knowledge recommendations. The experimental results show that after the introduction of the standard question bank, the accuracy of the intelligent recommendation system in specific industry knowledge domains significantly increased from 74.8% to 85.9%. Multiple sets of comparative experimental results further validate the effectiveness of the strategy of establishing a standard question bank in improving accuracy. The large model discussed in this paper optimized the intelligent knowledge recommendation for operator AI customer service, providing new ideas and technical support for the knowledge recommendation in telecommunications operators’AI customer service systems. With this model, operators can more effectively understand and respond to customer inquiries, significantly enhancing the customer service experience.

Current unsupervised contrastive learning methods mainly rely on pure textual information to construct sentence embeddings, which presents limitations in comprehensively understanding the deeper meanings conveyed by sentences. Meanwhile, traditional contrastive learning methods focus excessively on maximizing the mutual information between positive instances of text, overlooking the potential noise interference within sentence embeddings. To effectively retain useful information in the text while eliminating noise interference in the embeddings, the paper proposed a contrastive learning model based on text-vision and information entropy minimization. Firstly, the text and the corresponding visual information are deeply fused under the framework of contrastive learning, and jointly mapped to a unified grounding space, ensuring their representations remain consistent within this space. This approach overcomes the limitations of relying solely on pure textual information for sentence embedding learning, making the contrastive learning process more comprehensive and precise. Secondly, following the principle of information minimization, the model reconstructs positive text instances based on information entropy minimization while maximizing mutual information between positive text instances. Experimental results on the standard semantic textual similarity (STS) task demonstrate that the proposed model achieves significant improvements in the Spearman correlation coefficient evaluation metric, showcasing a notable advantage over existing state-of-the-art methods. This also confirms the effectiveness of the proposed model.

When using semantic segmentation methods to automatically segment barefoot footprint images, although manual intervention can be reduced, the issue of blurred toe regions in barefoot footprint image segmentation requires the neural network model to pay more attention to feature extraction from these areas. For barefoot footprint images with uneven lighting, the model can establish contextual relationships between the global and local regions of the footprint, using the feature information from the global region to enhance the feature expression of the uneven lighting areas, thereby improving the accuracy and robustness of image segmentation. To address this, this paper proposed a barefoot footprint segmentation method based on multi-granularity feature-region relationships. By using local region labels, the method enhances feature representation in the toe area, extracts multi-granularity features of footprints, and integrates them with global footprint features to improve segmentation performance in blurred areas. Meanwhile, spatial transformations were applied to both the original image and the footprint feature map, and a matrix multiplication approach was used to establish a barefoot region relationship matrix between them. This relationship matrix was then utilized to spatially modulate the global barefoot features, achieving feature enhancement. Furthermore, this paper constructed an in-the-wild barefoot footprint dataset consisting of 1 100 barefoot footprint images from 25 individuals and conducted experiments on four types of barefoot footprint images: blurred, unevenly illuminated, both blurred and unevenly illuminated, and normal. The results show that the intersection over union (IoU) for the barefoot class reaches 93.50% on normal barefoot footprint images. For blurred, uneven lighting, and blurry-uneven lighting images, the IoU are 92.90%, 93.06%, and 91.66%, respectively. Notably, the IoU for blurry-uneven lighting images is improved by 1.15 percentage points compared to U-Net.

Four-way Shuttle Automated Storage and Retrieval System (FWS-AS/RS) is a storage form widely used in industries such as e-commerce, pharmaceuticals, and food in recent years. It has the characteristics of flexible system configuration, high storage density, high efficiency, and high automation. To efficiently design the shelves layout and equipment configuration of FWS-AS/RS, this paper firstly considered the possibility of arranging shelf rows, columns, and layers under different storage scales and discussed the impact of changes in horizontal aisle layout position and quantity as well as different shelf depths on operational efficiency. Then, taking the different configurations of shuttle cars and lifts as well as the different placement positions of lifts and input/output (I/O) ports as variables, it established the motion models of shuttle cars and lifts considering acceleration, deceleration, no-load, load energy consumption, and energy recovery during braking; it used the total cost, transport distance, energy consumption, and space utilization rate as evaluation indicators. Through simulation experiments, regular design strategies were obtained for constructing four-way shuttle style stereoscopic warehouse shelves, including row, column, layer, depth, tunnel position, number and position of I/O ports, the ratio of four-way shuttle cars (FWS) to lifts, and the relationship between lifts and longitudinal tunnels. Finally, taking a storage capacity of 5 000 as an example, the study applied these strategies to design simulation. The simulation results show that the three optimization schemes reduced the transport distance, energy consumption, total cost, and floor area by an average of 43.30%, 57.69%, 11.17%, and 8.60%, while increasing the space utilization rate by an average of 5.66%, thus verifying the correctness of the design strategy. The design strategy can provide reference for the construction and operation of such stereoscopic warehouses.

Metal-oxide thin-film transistor (MO-TFT) can be utilized to create a flexible wearable system for detec-ting heart rate signals. However, the lack of high-performance complementary devices in MO-TFTs results in low gain for the implemented preamplifiers. Additionally, the relatively poor performance of MO-TFT devices poses challenges for the design of subsequent modules. In order to improve the gain of the preamplifier and reduce the performance requirements of the subsequent digital circuit, this paper proposed a common source common gate capacitor bootstrap structure preamplifier. The preamplifier was mainly composed of an external coupling bias module and a core amplifier module. The core amplifier module uses a capacitor bootstrapping technique known for its excellent stability, large output voltage swing, and low power consumption. This technique is combined with a cascode structure to enhance the overall gain of the circuit. The external coupling bias module utilizes an AC-coupled external bias structure that features low power consumption, high input impedance, and straightforward operating point setting, thereby meeting the bandpass requirements of the preamplifier for heart rate signal detection. The proposed preamplifier was designed and fabricated using a 10 μm IZO-TFT process. The test results indicate that with a 20 V power supply voltage, the circuit has a gain of 35 dB, a bandwidth of 2 Hz~2 kHz, a noise of 118.2 μV, and a power consumption of 0.1 mW. The presented preamplifier meets the requirements for detecting heart rate signals. In comparison to the current MO-TFT heart rate signal detection preamplifier, the gain has been increased by about 10 dB, which reduces the performance requirements of the subsequent digital module on the device, and is beneficial to achieve the digitalization of analog signals and maintain the signal integrity.

In the industrial Internet of Things, the reliability of mobile edge computing largely depends on the wireless channel conditions. In order to process the influence of imperfect channel state information to the system, this paper proposed a digital twin assisted mobile edge computing energy consumption optimization method. For the task offloading problem in industrial Internet of Things, a digital twin model of devices and channels in the edge computing system was established. Considering imperfect channel state information, the joint optimization of offloading decisions, transmission power, channel resources, and computational resources is performed with the aim of minimizing the total system energy consumption. To deal with the proposed nonlinear non convex problem of mixed integers, the probabilistic delay constraint was transformed and the original problem was decomposed into two sub-problems, and a joint optimization algorithm with the assistance of digital twins based on continuous convex approximation was proposed. Firstly, the original problem was relaxed to obtain resource allocation schemes and task offloading priorities. Then, the offloading priority of each terminal device was sorted in descending order. The complete task offloading scheme was obtained by solving the iterative optimization problem. Finally, simulation results show that, compared to other benchmark schemes, the proposed computational offloading optimization scheme significantly reduces the total energy consumption of the system.

Due to the susceptibility of TC4 material to oxidation failure in high-temperature environments, its service life is significantly shortened under harsh conditions such as high temperatures and marine environments. To augment the high-temperature oxidation resistance of TC4 material surfaces, this paper employed laser cladding technology to prepare a high-temperature oxidation-resistant cladding coating with a gradient mass fraction of additive phases on the TC4 surface. The microstructure of the coating was observed using scanning electron microscopy (SEM), and the impact of the additive phase on the microstructure morphology of the cladding was analyzed. Subsequently, microhardness tests were performed to obtain the microhardness distribution of coatings with different material compositions, and the effect of additive phase content on the microhardness of the cladding was analyzed. Ultimately, macroscopic morphology observation, oxidation kinetics, SEM, and XRD methods were employed to evaluate the high-temperature oxidation resistance of the cladded samples after high-temperature oxidation tests. The effects of ceramic phase content and the high-temperature oxidation process on the microstructure and phase composition of the cladding layer were analyzed, and the oxidation resistance mechanism of the coating was explored. The experimental findings reveal that the incorporation of ceramic phase powders results in a marked improvement in the microhardness of the cladding layer, along with a refinement and densification of its microstructure. The dense oxide products formed during the high-temperature oxidation process effectively isolate the coating from the oxidizing environment, thereby substantially enhancing its resistance to high-temperature oxidation. The high-temperature oxidation product Ta₂O₅ formed on the surface of the cladding layer has a dense structure, strong high-temperature stability, and excellent oxidation resistance, which is the main reason for the improved high-temperature oxidation resistance of the ceramic phase-containing TC4 cladding layer.

To address the unclear effects of quenching-polishing-quenching (QPQ) treatment on the performance of nickel-aluminum bronze alloy coatings welded onto 27SiMn alloy steel, this study investigated the geometric characteristics, microstructural changes, corrosion resistance, and hardness of the copper alloy coating. The influence of the QPQ treatment process on the coating’s microstructure and properties was analyzed to verify the rationality and feasibility of this composite anti-corrosion technology. The results indicate that after undergoing the two processes of carbonization/nitriding and oxidation, the copper alloy coating forms a dual-layer infiltration structure, with metal carbides distributed across both layers and copper oxides concentrated near the surface, providing corrosion protection. Before and after QPQ treatment, the microstructure of the copper alloy coatings primarily consists of the matrix phase α, the metastable phase, and various κ phases dispersed within the α phase. High-and medium-temperature tempering leads to the precipitation of a large amount of β' phase into the α phase, causing the matrix phase to coalesce and expand while reducing overall hardness. According to the protection rating representation method based on the proportion of substrate area affected by corrosion, the protection rating of the surface of the copper alloy coatings layer samples is 9, while the copper alloy samples after QPQ treatment is 10. The corrosion resistance of the copper alloy surface after QPQ treatment is not only maintained but also exceeds that of the untreated one. In light of this, the composite anti-corrosion technology can be applied to the maintenance and remanufacturing of hydraulic bracket cylinder barrel parts, which can enhance the corrosion resistance of the inner wall of the cylinder barrel as a whole, while also considering the corrosion resistance of other parts such as the joint holes and the outer surface of the cylinder body.

The recovery, storage and reuse of low-temperature waste heat in industry by using phase change materials for heat storage is an important method to achieve the gradual utilisation of energy and improve the efficiency of energy utilisation. The physical properties of phase change materials are the key factors determining the performance of heat storage systems.Therefore, the development of phase change materials with an appropriate phase transition temperature and good thermal cycling stability is of great significance for achieving efficient waste heat recovery. Based on this, a new phase change material, NaNO₃-KNO₃-NaNO₂-LiNO₃, was synthesized using the static melting method. A series of characterizations were conducted to evaluate its thermal properties, including melting point, latent heat, specific heat capacity, and cyclic stability, using differential scanning calorimetry, thermogravimetric analysis, X-ray diffraction, and Fourier transform infrared spectroscopy. The optimal composition was identified as m（NaNO₃）∶m（KNO₃）∶m（NaNO₂）∶m（LiNO₃）=6.32∶47.83∶36.10∶9.75 which was selected as the final preferred salt. The experimental results demonstrate that the preferred salt has significant performance advantages, with a low melting point of 79.02 ℃ and a latent heat of phase transition of 176.71 J/g; the average specific heat capacities of the solid and liquid phases are 1.96 and 2.09 J/(g·℃), respectively; the decomposition temperature reaches more than 600 ℃, which demonstrates its wide applicability in terms of temperature; after 100 high and low temperature cycling tests, the preferred salt still exhibited good thermal cycling stability. This study provides a new type of phase change energy storage material for low and medium temperature waste heat recovery and heat storage system, which is of great significance for energy optimisation and energy saving and emission reduction in related fields.

The copper smelting slag, abundant in valuable elements such as Fe and Si, exhibits excellent secondary resource characteristics and can be utilized as a raw material for constructing high-value-added silicon-iron-based functional materials. Understanding the controllable release patterns of Si and Fe elements under acid leaching conditions and the effective separation mechanisms of mineral phases is crucial for their high-value resource utilization.This study employed HSC 6.0 to simulate the dominant species in the silicon-iron system under varying pH and potential conditions, investigating the dissolution conditions of iron-containing mineral phases in the slag and the controllable release patterns of Si and Fe elements under H₂SO₄ acid leaching conditions. The effects of acid leaching temperature, H₂SO₄ concentration, particle size, and stirring speed on Fe leaching rate were analyzed. The results indicate that acid leaching temperature and H₂SO₄ concentration have a positive impact on the Fe leaching rate, while particle size exerts a negative influence, and stirring speed has minimal effect. Under conditions of 2.0 mol/L H₂SO₄ concentration, 90 ℃ acid leaching temperature, and copper slag particle size ranging from (45, 88]μm, the iron leaching rate can reach 95.73% after 60 minutes of acid leaching. The shrinking unreacted core model was used to describe the leaching process. In the initial stage of the reaction, the reaction rate is primarily controlled by the chemical reaction process, with an activation energy of 40.99 kJ/mol, and subsequently shifts to internal diffusion control, with an activation energy of 8.70 kJ/mol. During the chemical reaction control stage, the influence indices for H₂SO₄ concentration and copper slag particle size were calculated to be 0.558 and -0.759, respectively, thereby establishing the macrokinetic equation for the atmospheric pressure leaching of copper smelting slag with H₂SO₄.

LTA-type zeolites is considered a highly promising CO₂ capture material due to their excellent pore structures and high selectivity for CO₂ adsorption. This study presented a green and sustainable synthesis approach using industrial titanium-containing slag waste as a raw material to prepare Na-LTA zeolite precursors. Calcium loading was adjusted via a conventional liquid-phase ion exchange (LPIE) method to produce a series of xCa-LTA zeolites, specifically designed to enhance CO₂ adsorption performance. The adsorption properties of the xCa-LTA zeolites were systematically evaluated through dynamic adsorption experiments. Results show that xCa-LTA zeolites not only significantly enhances CO₂ capture capacity, but also exhibits excellent selectivity in CO₂/N₂ and CO₂/CH₄ separation processes, with the 0.05Ca-LTA sample demonstrating the most outstanding adsorption performance. Under conditions of 25 ℃ and 105 Pa, the CO₂ adsorption rate of 0.05Ca-LTA is 4.95 times that of Na-LTA, with a maximum adsorption capacity of 4.02 mmol/g. Kinetic analysis indicates that the CO₂ adsorption behavior of 0.05Ca-LTA follows a pseudo-second-order kinetic model, where the adsorption process is synergistically dominated by both physisorption and chemisorption. This synergy not only accelerated adsorption rates but also improved overall efficiency. After five adsorption/desorption cycles, 0.05Ca-LTA maintains highly efficient and stable adsorption performance, demonstrating excellent cyclic regeneration capability. This study follows the ecofriendly concept of “treating waste with waste” providing a new approach for the high-value utilization of solid waste while offering important theoretical and application potential for the synergistic optimization of CO₂ capture and environmental pollution control.