Accurate and real-time three-dimensional size estimation is often required for strawberries under agricultural field environments. However, the vision systems are typically restricted to the data capture from a single, oblique viewpoint. Only a partial three-dimensional point cloud can be generated with the practical constraints in the ridge-planting systems. The visible upper surface of the fruit can be represented without the full geometric information. Concurrently, a fundamental modeling challenge can also exist, due mainly to the natural morphology of a strawberry. A conical shape can be characterized by a truncated calyx base and a tapered tip, which is deviated substantially from the simple symmetric shapes, such as the ellipsoids commonly used in the conventional fitting. This geometric mismatch has been one of the primary sources of the estimation inaccuracies. Particularly, the critical dimensions (like the bottom diameter) cannot be directly observed from the available data perspective. In this study, an integrated computational framework was proposed to combine a lightweight and enhanced segmentation network with a purpose-specific geometric model designed to match the true morphology of the fruit. A four-stage pipeline was constructed for data completeness and model accuracy. 1) The robust instance segmentation was first performed by an improved YOLOv8s-seg network. The backbone was augmented with a custom C2f_Faster_EMA module in order to enhance the multi-scale feature extraction for the fine details with minimal computational overhead. 2) The high-precision mask was fused with the synchronized depth data to generate an initial 3D point cloud. The critical preprocessing involved the statistical outlier removal and voxel grid downsampling. Several thousand raw points were efficiently condensed into a consistent set of approximately 100 to 150 representative points, thereby preserving the key geometric features to reduce the computational load. 3) Subsequently, the hybrid RANSAC-PCA algorithm was operated on the refined point cloud to robustly estimate the strawberry's principal axis and 3D centroid. 4) The final stage employed a semi-truncated cone model, which was defined by the height, top diameter, and bottom diameter. The processed point cloud was used to fit the size estimation after optimization. The performance of the framework showed that the segmentation model was achieved in a mAP@0.5 of 96.9% and a ripe fruit segmentation accuracy of 98.3%, while there was a high inference speed of 159.3 frames per second. The size estimation shared high accuracy under varied conditions. Tests showed that the average relative errors of 1.2%, 0.7%, and 1.7% were obtained for the longitudinal diameter, top diameter, and bottom diameter, respectively, under different viewing angles. The relative errors of 1.8%, 1.2%, and 4.8% were found over a practical range of the capture heights from 30 to 70 cm after evaluations, indicating the consistent robustness. A multi-sample test was also performed on 22 standard strawberry specimens. The precision was further validated by the average value of errors of -0.13, 0.03, and -0.03 mm for the three dimensions, respectively, indicating all within the sub-millimeter range. Furthermore, the direct comparative analysis showed that the improved model significantly outperformed the baseline approaches. The absolute estimation error was also reduced by 50% and 70%, respectively, for the top and bottom diameters, compared with a single-view TSDF method. Crucially, an accurate estimate was successfully provided for the bottom diameter, rather than in the semi-elliptical models. The end-to-end pipeline also required an average processing time of 3.3s per frame, indicating the feasibility and strong potential. The real-time field deployment can be extended to more applications, such as the strawberry phenotyping, in-field grading, and precision yield prediction. The finding can also offer a practical solution using affordable RGB-D sensing hardware.

This study aims to improve the efficiency of durability testing, and then verify the harvesting machinery in laboratory scenarios. The accelerated editing was investigated on the load spectrum, according to the operating loads of combine harvester components as an example. The testing efficiency was further improved to reduce the costs for better consistency in the load spectrum loading between field and laboratory bench tests. Different wavelet components were selected to accelerate the editing using traditional wavelet decomposition. The threshold of pseudo-damage ratio was used to filter the wavelet components. The high contribution of damaged segments was identified in the wavelet components using the extreme difference method. Among them, the peak and valley values were extracted from the reconstructed wavelet components. Differences among adjacent extreme values were calculated after reconstruction. The threshold was then applied to discriminate the specific damage segments during extraction. An accelerated editing of the load spectrum was developed using optimal wavelet components. A case study was selected as the operating components of the feeding system in a combine harvester. The torque load was compared with the wavelet decomposition and time-domain damage retention. The results indicated that there was a greater variation in the signal frequency components and statistical features represented by different wavelet components, compared with the traditional acceleration editing of wavelet decomposition. Therefore, the negative impact of acceleration editing was also found in the indiscriminate identification of the damaged segment for each wavelet component. The acceleration editing was potentially improved to exclude the specific high-frequency wavelet components from the damage segment calculation. The wavelet components were then filtered, according to the pseudo-damage ratio between each wavelet component and the original load. There was a more precise and flexible identification of damage segments using extreme differences. Particularly for the boundaries of damage segments, compared with the time-domain window and the envelope damage segment identification. The periodicity of load cycles was achieved within the extracted load segments, in order to further enhance the accuracy of damage segment identification. The accelerated editing with optimized wavelet components was achieved in the more concise acceleration under the condition where the pseudo-damage retention ratios were nearly consistent. Taking the header auger load as an example under 95% and 98% pseudo-damage retention ratios, the acceleration values of the current wavelet decomposition were improved by 11.59 and 15.72 percental points, respectively, compared with the traditional. The acceleration editing with the optimized wavelet components also demonstrated better applicability to the load of drive shafts in the critical operating components of the feeding system. The signal compression ratios increased for the reel, header, and conveyor loads by 6.77, 12.44, and 20.19 percental points, respectively, under the relatively consistent levels of pseudo-damage retention, compared with the traditional. The better-accelerated editing of the load spectrum also exhibited better versatility for the different loads of operational components. The potential applications can also be found for the accelerated testing of structural durability scenarios in industrial field.