KZKai Zhu

Kai Zhu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kai Zhu has a total of 14 co-authors. Here are the top co-authors by number of shared publications:

  • Kaida Xu

    Zhejiang Marine Fisheries Research Institute, Zhoushan 316000, China

    2 shared publications
  • Fei Wang

    Key Laboratory of Fisheries Remote Sensing, Ministry of Agriculture and Rural Affairs, East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, China

    1 shared publication
  • Haozhan Wang

    Sansha Marine Environmental Monitoring Center Station, State Oceanic Administration, Haikou 570100, China

    1 shared publication
  • Kaida Xu

    Key Laboratory of Sustainable Utilization of Technology Research for Fishery Resources of Zhejiang Province, Zhejiang Marine Fisheries Research Institute, Scientific Observing and Experimental Station of Fishery Resources for Key Fishing Grounds, Ministry of Agriculture and Rural Affairs, China

    1 shared publication
  • Keji Jiang

    East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, China

    1 shared publication
  • Kexiang Lu

    School of Fishery, Zhejiang Ocean University, Zhoushan 316022, China

    1 shared publication

Latest Publications

  • Open Access

    Research on Fish Recognition in Complex Backgrounds Using ViT-Enhanced YOLOv11

    by Xiangshuo He, Shenglong Yang, Wei Wang, Kai Zhu, Shengmao Zhang, Yang Dai, Keji Jiang, Fei Wang

    To address the common challenges in fish recognition tasks under complex backgrounds, such as target overlap, occlusion, and chaotic spatial distribution, an improved YOLOv11 recognition model based on the Vision Transformer (ViT) is proposed. Traditional Convolutional Neural Networks (CNNs) and the YOLO series models are limited by their local receptive fields, making it difficult to capture global semantic correlations in dense and heavily occluded fish target detection, which often leads to feature confusion and false detections. By embedding ViT modules at the beginning of the Head and at the end of the Backbone of YOLOv11, the self-attention mechanism of ViT is leveraged to capture global dependencies in the image, re-integrate and enhance multi-scale features from the Backbone and Neck, thus constructing two improved ViT models. Comparative experiments are conducted on the FishRecognition-2025 dataset, which contains 955 high-resolution RGB images covering nine common coastal fish species across four categories: single fish species, multiple classes separated, slight overlap of multiple fish species, and severe overlap of multiple fish species. Under identical training strategies and evaluation metrics, the four models—original YOLOv11, traditional CNN, ViT-Head, and ViT-Backbone—are compared. The results show that the second improved ViT model (with ViT placed at the end of the Backbone) outperformed the first improved model (with ViT placed at the beginning of the Head) in terms of mAP50 and mAP50-95. Moreover, its overall accuracy across the four test data categories (single fish species, multiple classes separated, slight overlap of multiple fish species, and severe overlap of multiple fish species) surpassed that of YOLOv11, CNN, and the first ViT model. Although its accuracy in single fish species and multiple classes separated scenarios was slightly lower than that of the CNN model, it demonstrated significant advantages in scenarios with slight overlap of multiple fish species and severe overlap of multiple fish species. These findings validate the effectiveness of the ViT module in global feature modeling and adaptability to complex backgrounds, suggesting a promising technical direction for future real-time recognition in fishery field operations.

  • Open Access

    Structural Analysis and Functional Prediction of Gut Microbiota in Wild and Cultured Striped Knifejaw (Oplegnathus fasciatus)

    by Kai Zhu, Susu Zhang, Kaida Xu, Haozhan Wang

    Understanding the role of gut microbiota in fish health is crucial for optimizing aquaculture practices and ensuring sustainable fish populations. In this study, the diversity and compositional differences of intestinal microbiota were comparatively analyzed between wild and cultured striped knifejaw (Oplegnathus fasciatus Kroyer, 1845). Using high-throughput 16S rDNA sequencing and bioinformatics, an in-depth investigation of the gut microbiota in both populations was conducted. The results revealed that the number of intestinal bacterial sequences was significantly higher in the cultured population than in the wild population. The study included 16 individuals from the wild population and 38 individuals from the cultured population, with an average weight of 67.7 ± 12.4 g and 44.9 ± 16.8 g, respectively. Alpha diversity analysis indicated that intestinal microbiota species richness and diversity were both greater in the cultured O. fasciatus. Furthermore, significant differences were observed in the intestinal bacterial communities between the two populations, with Pseudomonadota, Verrucomicrobia, and Bacillota dominating in the cultured population, whereas Pseudomonadota overwhelmingly dominated in the wild population. Functional prediction analysis revealed differences between the intestinal microbiota in pathways related to genetic and environmental information processing, as well as metabolism. This study provides critical data for understanding the structure and function of intestinal microbial communities in O. fasciatus and offers a theoretical foundation for optimizing farming strategies to improve fish health and growth performance.


  • Open Access

    Heavy Metal Concentrations in Wild and Cultured Oplegnathus fasciatus from the East China Sea and Associated Health Risks

    by Kexiang Lu, Weiguo Qian, Kai Zhu, Kaida Xu

    In this study, a comparison of the concentrations of eight heavy metals (including Fe, Mn, Cu, Zn, Cr, As, Cd, and Hg) was conducted between wild and cultured Oplegnathus fasciatus. Significant differences in the concentrations of Zn, Cd, As, and Hg were observed between wild and cultured Oplegnathus fasciatus. The results showed that the mean Zn concentration was significantly higher in cultured Oplegnathus fasciatus (3.051 ± 0.738 mg/kg) when compared to its wild counterpart (2.512 ± 0.407 mg/kg). In contrast, the mean Cd concentration was found to be lower in the cultured Oplegnathus fasciatus (0.001 ± 0.0007 mg/kg) than in the wild ones (0.003 ± 0.003 mg/kg). Likewise, the wild samples demonstrated a higher mean As concentration (1.494 ± 0.659 mg/kg) than the cultured samples (0.594 ± 0.215 mg/kg). Lastly, it was noted that the mean Hg concentration was considerably higher in the cultured Oplegnathus fasciatus (0.042 ± 0.016 mg/kg) than in the wild specimens (0.014 ± 0.011 mg/kg). Pollution levels and health risks were evaluated using the single-factor pollution index (SFI), metal pollution index (MPI), and health risk assessment methods. The results showed that, for Cu, Zn, Cr, and Cd, both wild and cultured Oplegnathus fasciatus had SFI values below 1 compared to the marine organism quality standards. The MPI values for wild and cultured Oplegnathus fasciatus were 0.188 ± 0.051 and 0.172 ± 0.054, respectively, both far below the safety limit of 2 for pollution-free aquatic products. The Hazard Index (HI) for wild and cultured Oplegnathus fasciatus were below 1, indicating no health risks from long-term consumption. A discriminant analysis, based on Zn, Cd, As, and Hg concentrations, distinguished wild from cultured Oplegnathus fasciatus with a 96.0% accuracy, remaining stable at over 94.9% upon cross-validation. These findings accurately evaluate that there is no risk to human health from consuming Oplegnathus fasciatus, which is significant in safeguarding public health.


  • Open Access

    Discrimination of different forms and oceanic regions of purpleback flying squid (Sthenoteuthis oualaniensis) based on stable isotopes and fatty acid composition

    by Kai Zhu, Kaida Xu, Wenbin Zhu

    Introduction: Pelagic cephalopods play a key role in the Chinese food supply. The market value of seafood frequently fluctuates based on its geographical origin and species, making it imperative to trace the origin and distinguish forms to guarantee food quality and accurate information.Methods: In this study, we used biochemical tracers, including stable isotope analysis and fatty acid composition profiling, to trace the origin of purpleback flying squid (Sthenoteuthis oualaniensis) from the Indian Ocean and South China Sea. We measured the δ13C and δ15N values and fatty acid (FA) contents in samples from different forms of S. oualaniensis obtained from these two oceanic regions. We analyzed the feeding and nutritional differences among these populations and conducted discriminant analysis.Results and discussion: Significant regional and form differences were observed in both isotopic values and FA profiles in the muscle tissues of S. oualaniensis. Forms with larger sizes exhibited higher δ13C and δ15N values, potentially resulting from different feeding preferences and geographical environments. The discriminant analysis revealed that isotopic composition could effectively distinguish individuals from different oceanic regions, whereas FA composition was more effective in distinguishing different forms within the same oceanic region. The combination of isotopes and FAs could accurately discriminate S. oualaniensis of different forms and from different oceanic regions, achieving a discrimination accuracy of 100%. This study provides a reference for tracing the origin of pelagic cephalopods with complex form structures.