Rice spikelet size, spikelet length and spikelet width, are very important traits directly related to a rice crop’s yield. The accurate measurement of these parameters is quite significant in research such as breeding, yield evaluation and variety improvement for rice crops. Traditional measurement methods still mainly rely on manual labor, which is time-consuming, labor-intensive and error-prone. In this study, a novel method, dubbed the “SSM-Method”, based on convolutional neural network and traditional image processing technology has been developed for the efficient and precise measurement of rice spikelet size parameters on rice panicle structures. Firstly, primary branch images of rice panicles were collected at the same height to build an image database. The spikelet detection model using convolutional neural network was then established for spikelet recognition and localization. Subsequently, the calibration value was obtained through traditional image processing technology. Finally, the “SSM-Method” integrated with a spikelet detection model and calibration value was developed for the automatic measurement of spikelet sizes. The performance of the developed SSM-Method was evaluated through testing 60 primary branch images. The test results showed that the root mean square error (RMSE) of spikelet length for two rice varieties (Huahang15 and Qingyang) were 0.26 mm and 0.30 mm, respectively, while the corresponding RMSE of spikelet width was 0.27 mm and 0.31 mm, respectively. The proposed algorithm can provide an effective, convenient and low-cost tool for yield evaluation and breeding research.

The number of grains per sea rice panicle is an important parameter directly related to rice yield, and it is also a very important agronomic trait in research related to sea rice breeding. However, the grain number per sea rice panicle still mainly relies on manual calculation, which has the disadvantages of being time-consuming, error-prone, and labor-intensive. In this study, a novel method was developed for the automatic calculation of the grain number per rice panicle based on a deep convolutional neural network. Firstly, some sea rice panicle images were collected in complex field environment and annotated to establish the sea rice panicle image data set. Then, a sea grain detection model was developed using the Faster R-CNN embedded with a feature pyramid network (FPN) for grain identification and location. Also, ROI Align was used to replace ROI pooling to solve the problem of relatively large deviations in the prediction frame when the model detected small grains. Finally, the mAP (mean Average Precision) and accuracy of the sea grain detection model were 90.1% and 94.9%, demonstrating that the proposed method had high accuracy in identifying and locating sea grains. The sea rice grain detection model can quickly and accurately predict the number of grains per panicle, providing an effective, convenient, and low-cost tool for yield evaluation, crop breeding, and genetic research. It also has great potential in assisting phenotypic research.

Introduction: The detection of lucky bamboo (Dracaena sanderiana) nodes is a critical prerequisite for machining bamboo into high-value handicrafts. Current manual detection methods are inefficient, labor-intensive, and error-prone, necessitating an automated solution.Methods: This study proposes an improved YOLOv7-based model for real-time, precise bamboo node detection. The model integrates a Squeeze-and-Excitation (SE) attention mechanism into the feature extraction network to enhance target localization and introduces a Weighted Intersection over Union (WIoU) loss function to optimize bounding box regression. A dataset of 2,000 annotated images (augmented from 1,000 originals) was constructed, covering diverse environmental conditions (e.g., blurred backgrounds, occlusions). Training was conducted on a server with an RTX 4090 GPU using PyTorch.Results: The proposed model achieved a 97.6% mAP@0.5, significantly outperforming the original YOLOv7 (83.4% mAP) by 14.2%, while maintaining the same inference speed (100.18 FPS). Compared to state-of-the-art alternatives, our model demonstrated superior efficiency. It showed 41.5% and 153% higher FPS than YOLOv11 (70.8 FPS) and YOLOv12 (39.54 FPS), respectively. Despite marginally lower mAP (≤1.3%) versus these models, the balanced trade-off between accuracy and speed makes it more suitable for industrial deployment. Robustness tests under challenging conditions (e.g., low light, occlusions) further validated its reliability, with consistent confidence scores across scenarios.Discussion: The proposed method significantly improves detection accuracy and efficiency, offering a viable tool for industrial applications in smart agriculture and handicraft production. Future work will address limitations in detecting nodes obscured by mottled patterns or severe occlusions by expanding label categories during training.

Pearl detection with a counter (PDC) in a noncontact and high-precision manner is a challenging task in the area of commercial production. Additionally, sea pearls are considered to be quite valuable, so the traditional manual counting methods are not satisfactory, as touching may cause damage to the pearls. In this paper, we conduct a comprehensive study on nine object-detection models, and the key metrics of these models are evaluated. The results indicate that using Faster R-CNN with ResNet152, which was pretrained on the pearl dataset, mAP@0.5IoU = 100% and mAP@0.75IoU = 98.83% are achieved for pearl recognition, requiring only 15.8 ms inference time with a counter after the first loading of the model. Finally, the superiority of the proposed algorithm of Faster R-CNN ResNet152 with a counter is verified through a comparison with eight other sophisticated object detectors with a counter. The experimental results on the self-made pearl image dataset show that the total loss decreased to 0.00044. Meanwhile, the classification loss and the localization loss of the model gradually decreased to less than 0.00019 and 0.00031, respectively. The robust performance of the proposed method across the pearl dataset indicates that Faster R-CNN ResNet152 with a counter is promising for natural light or artificial light peal detection and accurate counting.

Foggy ship detection frequently suffers from image degradation, blurred object contours and a high missed detection rate. Moreover, most existing maritime datasets lack adequate real fog samples. To solve the above problems, in this paper, the Fog-SMD is constructed on the basis of atmospheric scattering principles and fractal theory in combination with a diffusion model to enrich samples covering various fog scenarios. On this basis, we develop an improved SF-YOLO model that takes YOLOv12 as the basic framework. By embedding the shallow–deep adaptive feature fusion module, scattering-guided refinement module and spatial-frequency dual feature attention module, the model can effectively alleviate feature loss resulting from image degradation in foggy environments. Weighted-EIoU loss is introduced to optimize the bounding box regression and reduce the localization deviation of slender ship targets. The experimental results show that SF-YOLO achieves mAP@50 and mAP@50:95 values of 79.3% and 61.6%, respectively, and outperforms mainstream detection algorithms; compared with YOLOv12n, it improves mAP@50:95 from 59.6% to 61.6%, with only a slight increase in parameters from 2.5 M to 2.8 M, providing a new solution for the practical deployment of detection systems and all-weather maritime monitoring in low-visibility foggy scenarios.