The Chinese furniture market has a high demand for wood floors. Manual defect detection in wood floors is inefficient and lacks stability in accuracy. It is necessary to conduct research on automatic defect detection in wood floors. To improve the accuracy of detecting small defects in wood floors, this paper proposed a new network based on YOLOv7. The new network is called the cascade center of gravity YOLOv7 (CCG-YOLOv7). This paper designed cascade efficient layer aggregation networks (C-ELAN), streamlined the CBS, replaced the ELAN with the C-ELAN, introduced the rapid supervised attention module to connect the backbone and head layers, and simplified the head layer of the YOLOv7 network. These methods improved the detection accuracy and speed for detecting small defects on wood floor surfaces. The improved network can effectively detect small defects on the wooden board surfaces, including knots, scratches, and mildew. Compared to the original YOLOv7, CCG-YOLOv7 improves precision, recall, and mean average precision by 2.1%, 1.6%, and 1.2%, respectively.

In the context of intelligent manufacturing for furniture, the classification of solid wood floors is critical for automating quality control and enhancing production efficiency. The final aesthetic of installed solid wood floors largely depends on the uniformity of color and texture. However, most current research does not address the simultaneous classification of both texture and color in solid wood floors, and there is a notable gap in studies focusing on the classification of color-difference wood floors and white-edge wood floors. This paper proposes a re-parameterized CNN-Transformer hybrid architecture based on the MobileViT model, named Rep-MobileViT. Specifically, the RepAIRB module is introduced, incorporating an asymmetric convolutional block (ACB) and a re-parameterized structure within the inverted residual block (IRB) module to enhance the network’s receptive field without increasing computational costs. Additionally, the RepMFF-MobileViT modu le is designed, utilizing a multi-feature fusion strategy and the RepTransformer structure to reduce the number of parameters while improving the network’s expressiveness and stability. In the preprocessing stage, a contrast-limited adaptive histogram equalization and brightness blending (CLAHE-LB) technique is employed to enhance image clarity, highlighting texture details while preserving original color information. Experimental results indicate that the proposed method achieves an overall classification accuracy of 94.98%, with a processing time of only 20.94 ms. Comparative analyses demonstrate that this approach outperforms existing machine learning methods by over 12% in classification accuracy.

With the advancement of automation in modern agriculture, the demand for intelligence in the post-picking sorting of fruits and vegetables is increasing. As a significant global agricultural product, the defect detection and sorting of tomato is essential to ensure quality and improve economic value. However, the traditional detection method (manual screening) is inefficient and involves high labor intensity. Therefore, a defect detection model named YOLO-RGDD is proposed based on YOLOv12s to identify five types of tomato surface defects (scars, gaps, white spots, spoilage, and dents). Firstly, the original C3k2 module and A2C2f module of YOLOv12 were replaced with RFEM in the backbone network to enhance feature extraction for small targets without increasing computational complexity. Secondly, the Dysample–Slim-Neck of the YOLO-RGDD was developed to reduce the computational complexity and enhance the detection of minor defects. Finally, dynamic convolution was used to replace the conventional convolution in the detection head in order to reduce the model parameter count. The experimental results show that the average precision, recall, and F1-score of the proposed YOLO-RGDD model for tomato defect detection reach 88.5%, 85.7%, and 87.0%, respectively, surpassing advanced object recognition detection algorithms. Additionally, the computational complexity of the YOLO-RGDD is 16.1 GFLOPs, which is 24.8% lower than that of the original YOLOv12s model (21.4 GFLOPs), facilitating the model’s deployment in automated agricultural production.

In panel-furniture manufacturing, dynamic scheduling is challenging because upstream motherboard release and parallel CUT-machine assignment are tightly coupled with downstream multibranch processing, finite-capacity buffers, and machine disturbances. This study investigates an event-driven scheduling problem in a multibranch panel-furniture workshop and aims to minimize the cumulative tardiness of all parts. To capture the structural dependencies of the shop floor, we construct a heterogeneous graph to represent the real-time production state. In this graph, candidate motherboards, parallel CUT machines, and downstream branch-level production units form different node types. The representation explicitly encodes motherboard–branch coupling features and machine–branch inflow relations, while aggregated branch-state features summarize detailed process-order information within each downstream branch. Based on this structured state representation, we develop a proximal policy optimization (PPO)-based actor–critic framework to learn adaptive motherboard release and CUT-assignment policies. We compare the proposed method with three heuristic rules and two PPO-based learning baselines under different machine-availability levels, buffer-capacity settings, and dataset scales. The results show that the proposed method achieves the best overall tardiness performance across the tested baselines and converges faster and more stably than the other PPO-based methods under different dataset scales and workshop conditions. These findings demonstrate the effectiveness of combining heterogeneous graph modeling with reinforcement learning for dynamic scheduling in multibranch panel-furniture production systems.

Freshness index is one of crucial factors that affect demand of perishable products. Since freshness index decays over time with uncertainty, freshness-keeping investment is thus an important decision for supply chains except for selling price. We investigate cooperation problem of dynamic pricing and freshness-keeping investment in a two-echelon perishable supply chain consisting of a distributor and one retailer. The retailer first announces a selling price. Additionally, he also shows a participation rate on freshness-keeping investment that specifies the fraction of the distributor’s freshness-keeping costs that he will support. In response, the distributor determines his optimal wholesale price and freshness-keeping policy. We propose a stochastic model to describe the dynamics of freshness index and formulate the coordination problem as a Stackelberg differential game. Applying stochastic dynamic programming, two types of closed-form equilibrium are analytically obtained. Moreover, we conclude the system condition when the retailer chooses co-op freshness-keeping equilibrium. The results are compared with the corresponding policies in the vertically integrated channel, and we demonstrate that a combination of revenue and freshness-keeping sharing contract can coordinate the chain. The study presents managers with a quantitative instrument to design freshness-keeping coordination.