Efficient large-scale 3D reconstruction of orchard environments is essential for robotic inspection and precision agriculture, yet existing methods struggle with unstructured scenes, variable illumination, and computational bottlenecks. We propose InspectGaussian, a coarse-to-fine Gaussian reconstruction framework tailored for orchard inspection robots. The pipeline integrates an RGB-D-based data acquisition strategy using ORB-SLAM3, which is enhanced by a dense mapping module for robust large-scale pose estimation and point cloud generation. A divide-and-conquer strategy is then employed: individual plant views are extracted via a YOLO-World–based detection and 3D matching algorithm, followed by plant-specific reconstruction using an improved 3D Gaussian Splatting (3DGS) method incorporating depth regularization and region-aware refinement. Experimental results in citrus orchards demonstrate that InspectGaussian achieves 96% average precision and 93% recall in plant view extraction, while surpassing state-of-the-art methods in reconstruction fidelity (31.226 PSNR, 0.915 SSIM, 0.067 LPIPS) and point cloud accuracy (7 mm error). These results confirm its effectiveness in capturing fine structural and textural details while maintaining scalability and efficiency. This framework provides a practical solution for high-throughput, in-field plant phenotyping and lays the foundation for intelligent orchard monitoring and management.

3D phenotyping of the external and internal structures is important to breed new fruit species. As manual phenotyping is error-prone and time-consuming, developing high-throughput solutions with enhanced precision and low costs is necessary. This study presents CitrusGAN, a generative adversarial network-based method to reconstruct 3D citrus CT models from sparse-view X-ray images. The input X-rays are arranged in orthogonal pairs to provide additional information, and customized loss functions enable more effective learning of the mapping from 2D X-ray features to 3D CT volumes. Experimental results show that 6 views can generate high-quality citrus CT volumes, with a structural similarity index of 92.1 ​% and a peak signal-to-noise ratio of 26.374 ​dB compared with the real CT models. Moreover, the morphology of the generated model can be conveniently measured in the 3D space, facilitating the extraction of phenotypic traits including fruit length, width, height, volume, surface area, peel thickness, number of segments, and edible rate with high precision. As X-rays can be obtained using low-cost X-ray machines with high efficiency, the proposed method can be potentially developed into high-throughput equipment for fruit production lines or portable devices to realize in-field phenotyping.

Citrus rind color is a good indicator of fruit development, and methods to monitor and predict color transformation therefore help the decisions of crop management practices and harvest schedules. This work presents the complete workflow to predict and visualize citrus color transformation in the orchard featuring high accuracy and fidelity. A total of 107 sample Navel oranges were observed during the color transformation period, resulting in a dataset containing 7,535 citrus images. A framework is proposed that integrates visual saliency into deep learning, and it consists of a segmentation network, a deep mask-guided generative network, and a loss network with manually designed loss functions. Moreover, the fusion of image features and temporal information enables one single model to predict the rind color at different time intervals, thus effectively shrinking the number of model parameters. The semantic segmentation network of the framework achieves the mean intersection over a union score of 0.9694, and the generative network obtains a peak signal-to-noise ratio of 30.01 and a mean local style loss score of 2.710, which indicate both high quality and similarity of the generated images and are also consistent with human perception. To ease the applications in the real world, the model is ported to an Android-based application for mobile devices. The methods can be readily expanded to other fruit crops with a color transformation period. The dataset and the source code are publicly available at GitHub.