Zhipeng He

Mr. Zhipeng He

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 May 2026 in Agriculture
Verified
Journal Article
Open access
PDF file
YSYifei Sun
Yifei Sun
XMXinying Miao
Xinying Miao
YSYifei Sun
Yifei Sun
ZHZhipeng He
Zhipeng He
XTXinyue Tao
Xinyue Tao
Wei WangWei Wang
Wei Wang
Sweet cherry cracking severely impairs its commercial value and causes huge economic losses, and the accurate real-time detection of fine cracking defects remains a challenging small-target detection task. Traditional manual sorting and conventional machine vision methods suffer from low efficiency and poor robustness, while existing YOLO-based models have limitations in multi-scale feature fusion, local feature discrimination and spatial information retention for cherry cracking detection, and their effectiveness in natural production environments has not been statistically validated. To address these issues, this study proposes YOLO-CY for cherry cracking defect detection. Three key modules were optimized: the C3k2_AdditiveBlock was designed to enhance multi-scale feature extraction, the C2PSA_CGLU module improved the discriminability of local crack features via refined channel attention, and the Efficient Up-Convolution Block replaced traditional upsampling to reduce spatial information loss. Experiments were conducted on a self-constructed dataset of 3662 cherry images acquired on a real sorting line under natural ambient light. The results showed that YOLO-CY achieved an mAP50 of 94.88% and an mAP50-95 of 64.92%, with precision and recall reaching 93.90% and 90.81%, respectively, significantly outperforming mainstream lightweight YOLO models and two-stage detectors. Ablation experiments verified the synergistic effect of the three improved modules, and the model only had a marginal increase in parameters (2.62 M) and GFLOPs (6.60), maintaining lightweight characteristics. YOLO-CY can accurately detect fine, low-contrast and pedicel-overlapping cracks and is suitable for real-time detection on automated cherry-sorting lines, providing a technical solution for intelligent cherry quality inspection.

6 October 2026 in Processes
Verified
Journal Article
Open access
PDF file
MLMenghan Liu
Menghan Liu
XMXinying Miao
Xinying Miao
PRPing Ren
Ping Ren
JSJingjing Shan
Jingjing Shan
JLJunhao Li
Junhao Li
JZJiankang Zhong
Jiankang Zhong
Reliable fault diagnosis of rolling bearings is essential for ensuring the safe and stable operation of rotating machinery. However, existing diagnostic models are often hindered by the strong non-stationarity and weak fault characteristics of vibration signals under complex operating conditions, as well as uneven sample distributions among fault categories, which compromise discriminative representation learning and lead to biased fault recognition. To address these challenges, this paper proposes a novel rolling bearing fault diagnosis framework, termed MOBO-VMD-SPD-ProSwinNet, which integrates adaptive signal decomposition, geometry-aware fault representation, and semantic-guided deep feature learning. Specifically, multi-objective Bayesian optimization (MOBO) is introduced to adaptively determine the key parameters of variational mode decomposition (VMD), while the selected high-energy intrinsic mode functions are transformed into Riemannian symmetric positive definite (SPD) recurrence images to preserve local statistical dependencies and temporal recurrence characteristics. On this basis, a convolutional neural network (CNN)-Swin Transformer architecture is combined with prompt-guided semantic prototype learning to capture complementary local–global fault information while enhancing class-oriented representation and reducing class-biased feature learning. Experiments on the Case Western Reserve University (CWRU), Jiangnan University (JNU), and Southeast University (SEU) bearing datasets demonstrate the effectiveness of the proposed framework. On the CWRU benchmark, the proposed method achieves mean Accuracy, Macro-Precision, Macro-Recall, and Macro-F1 of 99.59%, 99.61%, 99.60%, and 99.59%, respectively, improving the mean accuracy by 1.75 percentage points over the strongest evaluated comparison method. Moreover, mean accuracies of 99.64% and 98.86% are obtained on the JNU and SEU datasets, respectively, further demonstrating the robustness of the proposed framework across different operating conditions and data distributions.

13 October 2026 in Sensors
Verified
Journal Article
Open access
PDF file
JSJingjing Shan
Jingjing Shan
XMXinying Miao
Xinying Miao
MLMenghan Liu
Menghan Liu
HLHaoran Li
Haoran Li
ZHZhipeng He
Zhipeng He
HWHan Wang
Han Wang
Routine microalgal cell counting requires sampling and microscopy. We developed a low-cost portable machine-vision system (LPMVS) for non-destructive density estimation of visibly green Dunaliella sp. cultures. The compact benchtop LPMVS combines motorized flask loading, a light-isolated dark box with diffuse light-emitting diode (LED) illumination, fixed camera–flask geometry, red–green–blue (RGB) imaging through the flask, mask-derived liquid-region extraction, Raspberry Pi inference, and a local liquid-crystal display (LCD). A ready-state load-to-unload cycle required approximately 37 s, including 6 s of processing. The two 40-day flask trajectories yielded 240 image–hemocytometer pairs; 200 records were used to compare three regressors with 20 features selected from 339 descriptors. K-nearest neighbors (KNN) performed best in the flask-mixed record-level comparison (R2 = 0.98; mean absolute percentage error (MAPE) = 6.41%; n=40). Reciprocal leave-one-flask-out transfer analysis on complete trajectories with independently retrained pipelines gave KNN R2 values of 0.914 and 0.803. In the reverse direction, R2 was 0.936 within the overlapping range (n=79), but 41 above-range densities were systematically underestimated. Bootstrap resampling identified recurrent color-based features in both trajectories. An edge-device check using 10 additional measurements yielded preliminary reference-density tracking (R2 = 0.94). These findings support the feasibility of the LPMVS workflow for non-destructive monitoring under controlled laboratory conditions, but extrapolation and generalization to independent batches and optical configurations remain unestablished.