Today, different models and instruments exist to study and model color vision and color vision deficiency. These systems are often modeled at spectral and retinal levels. In this study, we propose a novel approach to set up models and aids for color vision deficiencies, considering the role of spatial color processing in human visual system. In particular, we present the results of a perceptual test to identify the role of the spatial arrangement in color discrimination by Color Deficient Observers (CDOs) and Color Normal Observers (CNOs), using simultaneous contrast effect.
The rapid evolution of modern society has triggered a surge in the production of diverse waste in daily life. Effective implementation of waste classification through intelligent methods is essential for promoting green and sustainable development. Traditional waste classification techniques suffer from inefficiencies and limited accuracy. To address these challenges, this study proposed a waste image classification model based on DenseNet-121 by adding an attention module. To enhance the efficiency and accuracy of waste classification techniques, publicly available waste datasets, TrashNet and Garbage classification, were utilized for their comprehensive coverage and balanced distribution of waste categories. 80% of the dataset was allocated for training, and the remaining 20% for testing. Within the architecture of DenseNet-121, an enhanced attention module, series-parallel attention module (SPAM), was integrated, building upon convolutional block attention module (CBAM), resulting in a new network model called dense series-parallel attention neural network (DSPA-Net). DSPA-Net was trained and evaluated alongside other CNN models on TrashNet and Garbage classification. DSPA-Net demonstrated superior performance and achieved accuracies of 90.2% and 92.5% on TrashNet and Garbage classification, respectively, surpassing DenseNet-121 and alternative image classification algorithms. These findings underscore the potential for executing efficient and accurate intelligent waste classification.
To address the challenges in chip logo detection, such as the small size of the logos making them difficult to be detected accurately and the slow convergence speed of traditional models, we propose a real-time detection algorithm for small objects, called small-DETR. First, to reduce production costs and enhance efficiency, we employ a semi-automated data annotation method based on template matching instead of traditional manual annotation, generating label files for model training and testing. Subsequently, building upon the RT-DETR algorithm, we enhance the feature fusion module in cross-scale feature fusion module (CCFM) using semantics and details injection (SDI) module from U-Net v2. This improvement aims to retain detailed image information, accurately capturing edges, textures, and subtle variations within the marks. Lastly, employing FasterNet as the backbone network for the detection model, we optimize the existing network structure using partial convolution (PConv) to reduce redundant computations and improve convergence speed. Experimental results demonstrate that small-DETR model achieves satisfactory convergence in just 200 cycles, with a detection precision of 91.8% and a loss value of 6.1%. Compared to other models, small-DETR exhibits outstanding performance within shorter training periods, providing robust support for real-time chip pin mark detection in industrial contexts.
In the smart manufacturing process, it is important to closely monitor manufactured parts. To solve the problem of part anomaly detection, this paper proposes a GAM–Boost anomaly detection model using a large-scale dataset (14.3 GB) from the Kaggle competition “Bosch Production Line Performance.” The model first selects the important features using the XGBoost algorithm and then captures the nonlinear relationships between the features using the generalized additive model. To capture the nonlinear relationships between features and at the same time improve the model’s ability to understand the data relationships, feature engineering techniques are applied to transform the nonlinear relationships without ignoring the linear relationship features. Finally the XGBoost model is optimized for anomaly detection using the Bayesian algorithm. The experimental results show that the model achieves lower errors on both training and test sets, the generalization performance of the model is significantly improved, it can better adapt to various data situations, and it achieves better results in terms of flexibility and prediction accuracy.
Ensuring smooth tension is imperative to maintaining the quality of lithium battery coating; particularly, the precision of tension control within the unwinding system is paramount to ensuring high-quality outcomes in subsequent processes and product fabrication. This paper proposes a fuzzy sliding mode variable structure control strategy based on the tension control of the unwinding system to meet the stability requirements of tension in the lithium battery coater’s unwinding system. First, we establish a nonlinear time-varying dynamic model of the unwinding system by elucidating the operational principles of the lithium battery coater unwinding system. Second, leveraging the tension model of the unwinding system, we devise a fuzzy sliding mode variable structure controller specifically tailored for tension control, subsequently conducting a stability analysis of the system. Finally, the effectiveness of the proposed fuzzy sliding mode variable structure controller is validated through simulation tests. Experimental results demonstrate that the devised fuzzy sliding mode variable structure controller exhibits superior robustness compared to both traditional proportional–integral–derivative control and sliding mode variable structure control.
Unmanned aerial vehicles (UAV) are extensively utilized in various applications due to their compact size and flexibility. However, the detection task in UAV images faces significant challenges stemming from the abundance of small targets and the wide range of target sizes. To address this issue, we propose an object detection method specifically designed for UAV images, incorporating two enhancements to the strong baseline model YOLOX, which excels at detecting small and multi-scale targets. First, we introduce a high-resolution feature map to preserve detailed information crucial for small targets. We then introduce an attention mechanism to guide the model to focus more on small targets in high-dimensional features. Experimental results on the VisDrone-VID2019 dataset confirm the effectiveness and superiority of our proposed method.
The 3D extension of the High Efficiency Video Coding (3D-HEVC) standard has improved the coding efficiency for 3D videos significantly. However, this improvement has been achieved with a significant rise in computational complexity. Specifically, the encoding process for the depth map in the 3D-HEVC standard occupies 84% of the total encoding time. This extended time is primarily due to the need to traverse coding unit (CU) depth levels in depth map encoding to determine the most suitable CU size. Acknowledging the evident texture distribution patterns within a depth map and the strong correlation between encoding size selection and the texture complexity of the current encoding block, an adaptive depth early termination convolutional neural network, named ADET-CNN, is designed for the depth map in this paper. It takes an original 64 × 64 coding tree unit (CTU) as the input and provides segmentation probabilities for various CU sizes within the CTU, which eliminates the need for exhaustive calculations and the comparison for determining the optimal CU size, thereby enabling faster intra-coding for the depth map. Experimental results indicate that the proposed method achieves a time saving of 58% depth map encoding while maintaining the quality of synthetic views.
Layout design is an important step in packaging graphic design, and high-quality layout design is an important attribute to attract consumers’ attention and subsequent purchase. In order to solve the problems of time-consuming, difficult communication and strong dependency faced by manually performing package graphic design, we propose a template-free package layout generation method to achieve intelligent layout design. This method uses generative adversarial network (GAN) as a framework, and the generator and discriminator are composed of two improved transformer structures, which combines the advantages of two mainstream generative models, taking into account the rich layout variations in packaging design for generating a robust layout. We also constructed a packaging dataset PackageLayout to verify the superiority of the proposed method, which contains 2020 packaging planar images and annotation information for three categories. After ablation experiments on the homemade packaging dataset and comparison experiments with current state-of-the-art methods (CGLGAN, DSGAN), we validated the effectiveness and stability of the model. The layouts generated by our model are visually similar to the real design layouts and outperform previous models in terms of evaluation metrics. Finally, we also constructed real designs based on the predicted layouts to better understand the visual quality, which contributes to the advancement of the application of intelligent layout design models in packaging graphic design.
With the development of China’s new economy, the express industry has also shifted from high-speed development to high-quality development. Joint distribution aims to reduce the repeated waste of resources caused by the separate transportation of various express delivery enterprises by means of joint transportation, and promote the transformation and upgrading of the express delivery industry to high-quality development. Therefore, the purpose of this paper is to effectively reduce the cost of and improve the efficiency of express delivery. In this paper, a dual-objective joint distribution network planning model with the highest efficiency and the lowest total cost of joint distribution network is established, which is solved by a three-stage algorithm. The feasibility of the model and algorithm has been applied for validation in the city of Beijing, and the reasonable planning of the express network provides an effective decision-making basis.
The current high-fidelity printing uses RGB images as original patterns, but the digital screening color matching program has adopted “CMYK+ spot color printing” method, which has a narrow color gamut and the phenomenon of metameric colors. Based on these problems, this study evaluated multispectral images as an image pattern for high-fidelity printing. However, multispectral images have too many color dimensions and information redundancy, which could not be used directly for printing. This study adopted the nonnegative principal component analysis method to reduce the dimensions of multispectral images to obtain the principal components of the colors, and the corresponding printing colors were selected. Finally, based on variable frequency amplitude modulation screening theory and its technical application rules, a new digital screening color matching program is proposed. Thus, more printing color inks could be used in high-fidelity printing output. Experimental results have shown that the new printing output method expanded the printing color gamut and reduced the degree of metamerism of the printed images. This novel idea and technical approach to achieve the high-fidelity printing has a significant theoretical and application value.