Choosing the optimal path in planning is a complex task due to the numerous options and constraints; this is known as the trip design problem (TTDP). This study aims to achieve path optimization through the weighted sum method and multi-criteria decision analysis. Firstly, this paper proposes a weighted sum optimization method using a comprehensive evaluation model to address TTDP, a complex multi-objective optimization problem. The goal of the research is to balance experience, cost, and efficiency by using the Analytic Hierarchy Process (AHP) and Entropy Weight Method (EWM) to assign subjective and objective weights to indicators such as ratings, duration, and costs. These weights are optimized using the Lagrange multiplier method and integrated into the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) model. Additionally, a weighted sum optimization method within the Traveling Salesman Problem (TSP) framework is used to maximize ratings while minimizing costs and distances. Secondly, this study compares seven heuristic algorithms—the genetic algorithm (GA), particle swarm optimization (PSO), the tabu search (TS), genetic-particle swarm optimization (GA-PSO), the gray wolf optimizer (GWO), and ant colony optimization (ACO)—to solve the TOPSIS model, with GA-PSO performing the best. The study then introduces the Lagrange multiplier method to the algorithms, improving the solution quality of all seven heuristic algorithms, with an average solution quality improvement of 112.5% (from 0.16 to 0.34). The PSO algorithm achieves the best solution quality. Based on this, the study introduces a new variant of PSO, namely PSO with Laplace disturbance (PSO-LD), which incorporates a dynamic adaptive Laplace perturbation term to enhance global search capabilities, improving stability and convergence speed. The experimental results show that PSO-LD outperforms the baseline PSO and other algorithms, achieving higher solution quality and faster convergence speed. The Wilcoxon signed-rank test confirms significant statistical differences among the algorithms. This study provides an effective method for experience-oriented path optimization and offers insights into algorithm selection for complex TTDP problems.

Hyperspectral image (HSI) classification is fundamental to a wide range of remote sensing applications, such as precision agriculture, environmental monitoring, and urban planning, because HSIs provide rich spectral signatures that enable the discrimination of subtle material differences. Deep learning approaches, including Convolutional Neural Networks (CNNs), Graph Convolutional Networks (GCNs), and Transformers, have achieved strong performance in learning spatial–spectral representations. However, these models often face difficulties in jointly modeling long-range dependencies, fine-grained local structures, and non-Euclidean spatial relationships, particularly when labeled training data are scarce. This paper proposes a Spectral–Spatial Graph Transformer Network (SSGTN), a dual-branch architecture that integrates superpixel-based graph modeling with Transformer-based global reasoning. SSGTN consists of four key components, namely (1) an LDA-SLIC superpixel graph construction module that preserves discriminative spectral–spatial structures while reducing computational complexity, (2) a lightweight spectral denoising module based on $1\times 1$ convolutions and batch normalization to suppress redundant and noisy bands, (3) a Spectral–Spatial Shift Module (SSSM) that enables efficient multi-scale feature fusion through channel-wise and spatial-wise shift operations, and (4) a dual-branch GCN-Transformer block that jointly models local graph topology and global spectral–spatial dependencies. Extensive experiments on three public HSI datasets (Indian Pines, WHU-Hi-LongKou, and Houston2018) under limited supervision (1% training samples) demonstrate that SSGTN consistently outperforms state-of-the-art CNN-, Transformer-, Mamba-, and GCN-based methods in overall accuracy, Average Accuracy, and the $\kappa$ coefficient. The proposed framework provides an effective baseline for HSI classification under limited supervision and highlights the benefits of integrating graph-based structural priors with global contextual modeling.

Sugar apple (Annona squamosa) is prized for its excellent taste, rich nutrition, and diverse uses, making it valuable for both fresh consumption and medicinal purposes. Predominantly found in tropical regions of the Americas and Asia, its harvesting remains labor-intensive in orchard settings, resulting in low efficiency and high costs. This study investigates the use of computer vision for sugar apple instance segmentation and introduces an improved deep learning model, GCE-YOLOv9-seg, specifically designed for orchard conditions. The model incorporates Gamma Correction (GC) to enhance image brightness and contrast, improving target region identification and feature extraction in orchard settings. An Efficient Multiscale Attention (EMA) mechanism was added to strengthen feature representation across scales, addressing sugar apple variability and maturity differences. Additionally, a Convolutional Block Attention Module (CBAM) refined the focus on key regions and deep semantic features. The model’s performance was evaluated on a self-constructed dataset of sugar apple instance segmentation images captured under natural orchard conditions. The experimental results demonstrate that the proposed GCE-YOLOv9-seg model achieved an F1 score (F1) of 90.0%, a precision (P) of 89.6%, a recall (R) level of 93.4%, a mAP@0.5 of 73.2%, and a mAP@[0.5:0.95] of 73.2%. Compared to the original YOLOv9-seg model, the proposed GCE-YOLOv9-seg showed improvements of 1.5% in the F1 score and 3.0% in recall for object detection, while the segmentation task exhibited increases of 0.3% in mAP@0.5 and 1.0% in mAP@[0.5:0.95]. Furthermore, when compared to the latest model YOLOv12-seg, the proposed GCE-YOLOv9-seg still outperformed with an F1 score increase of 2.8%, a precision (P) improvement of 0.4%, and a substantial recall (R) boost of 5.0%. In the segmentation task, mAP@0.5 rose by 3.8%, while mAP@[0.5:0.95] demonstrated a significant enhancement of 7.9%. This method may be directly applied to sugar apple instance segmentation, providing a promising solution for automated sugar apple detection in natural orchard environments.

Conventional fertigation systems suffer from limited cross-regional adaptability, mainly due to unstable fertilizer flow from fixed-aperture units, poor terrain adaptability, and an inadequate response to environmental heterogeneity. This study proposes an autonomous, cross-regional intelligent fertigation system based on an STM32F1 microcontroller and UART communication protocols. The system integrates a mechanically adjustable iris fertilizer delivery unit, a dual-axis fertigation module, a data interconnection unit, and comprehensive control software with dynamic calibration capabilities. Prototype evaluations conducted on both sloped terrain (up to 38°) and flat surfaces demonstrate a stable performance, achieving fertilizer flow control errors below 3%, irrigation deviation under 5%, and fertilization deviation within 2%. Real-time data acquisition, remote monitoring, and intelligent operation are supported by a YOLOv5s-based visual recognition system, which attains an mAP@0.5 of 92.5%. This integrated solution offers a robust approach to precision agriculture across diverse environmental conditions.