Chinese Journal of Ship Research, Volume. 20, Issue 2, 140(2025)

Operation standardization evaluation method based on improved YOLOv8n for ship equipment disassembly and assembly

Zhendong ZHANG1, Cong GUAN1, Zehui ZHANG2, Chao WU1, and Xuewen DING1
Author Affiliations
  • 1School of Naval Architecture, Ocean and Energy Power Engineering, Wuhan University of Technology, Wuhan 430063, China
  • 2School of Automation (School of Artificial Intelligence), Hangzhou Dianzi University,Hangzhou 310018, China
  • show less
    Figures & Tables(12)
    [in Chinese]
    [in Chinese]
    [in Chinese]
    [in Chinese]
    [in Chinese]
    • Table 1. Evaluation of membership grade table

      View table
      View in Article

      Table 1. Evaluation of membership grade table

      评价评估隶属度
      20分40分60分80分100分
      非常熟练0000.20.8
      熟练000.20.80
      一般00.20.60.20
      00.80.200
      较差0.80.2000
    • Table 2. Pseudo-code for the overall algorithm of this paper

      View table
      View in Article

      Table 2. Pseudo-code for the overall algorithm of this paper

      Algorithm 1: 拆装流程规范性评估算法
      Input: 拆装视频Doperation
      Output: 零件拆装流程评分G'
      1: Initialization
      2: (1) 确定拆装流程评估的关键因素集及评价集B
      3: (2) 初始化目标检测模型
      4: Procedure
      5: (1)设备拆装流程识别
      6: for i = 1 to n do
      7:  Obji←Detect(Improved YOLOv8n, Doperation(i))//识别零件i
      8:  Ooperation(i)←Determine{Obji, Objpump}//辨识拆装顺序
      9: end
      10: (2)拆装流程评估
      11: W←AHP&EWM//确定拆装关键因素权重集
      12: for i = 1 to n do
      13: R←Estimate{ Ooperation(i), Bi}//根据评价集得到隶属度矩阵
      14: end

      15: G'U×[20 40 60 80 100]TW×R//根据式(11)与式(12)得到评价结果并转化为百分制

      16: Return G'
    • Table 3. Comparative experiments of different attentional mechanisms

      View table
      View in Article

      Table 3. Comparative experiments of different attentional mechanisms

      模型评价指标
      PrReMap @0.5Pa×106Gf×109
      YOLOv8n[24]0.7400.760.803.018.1
      v8n-CBAM[31]0.8900.790.883.128.2
      v8n-CoTA[32]0.8900.600.814.0010.1
      v8n-CA[33]0.9000.790.853.028.1
      v8n-MHSA[34]0.8010.770.823.018.1
      本文SA注意力0.9100.630.933.048.3
    • Table 4. Ablation experiment

      View table
      View in Article

      Table 4. Ablation experiment

      模型评价指标
      PrReMap @0.5Pa×106Gf×109
      v8n[24]0.740.760.803.018.1
      v8n-S0.910.630.933.058.3
      v8n-W0.710.820.783.018.1
      v8n-G0.830.770.802.988.1
      v8n-SW0.820.810.943.058.3
      v8n-SG0.780.790.943.038.3
      v8n-GW0.920.780.882.988.1
      本文方法0.920.810.953.038.3
    • Table 5. Comparative experiment

      View table
      View in Article

      Table 5. Comparative experiment

      模型评价指标
      PrReMap @0.5Pa×106Gf×109
      YOLOv5n[35]0.840.810.791.774.2
      YOLOv5s[35]0.910.770.787.0315.8
      YOLOv7[36]0.930.780.7936.5104.7
      YOLOv8n[24]0.740.760.803.018.1
      YOLOv8s[24]0.710.810.8011.228.6
      YOLOv8x[24]0.830.760.7868.2257.8
      本文方法0.920.810.953.038.3
    • Table 6. Disassembly and assembly process identification experiment

      View table
      View in Article

      Table 6. Disassembly and assembly process identification experiment

      正确流程视频动作动作时间帧序号实时检测视频动作动作时间帧序号
      拆卸端盖3 676拆卸端盖3 730
      拆卸1个轴承套3 898拆卸1个轴承套4 013
      拆卸主动轴4 045拆卸主动轴4 172
      拆卸从动轴4 298拆卸从动轴4 196
      拆卸2个轴承套4 366拆卸2个轴承套4 426
      安装1个轴承套4 415安装1个轴承套4 679
      安装主动轴4 572安装从动轴4 789
      安装从动轴4 662安装主动轴4 803
      安装2个轴承套4 848安装2个轴承套4 931
      安装端盖5 002安装端盖5 103
    • Table 7. Historical disassembly data

      View table
      View in Article

      Table 7. Historical disassembly data

      样本编号对应指标
      A1A2A3A4A5
      1367916311647
      2373012610307
      3358215111453
      4366514801525
    Tools

    Get Citation

    Copy Citation Text

    Zhendong ZHANG, Cong GUAN, Zehui ZHANG, Chao WU, Xuewen DING. Operation standardization evaluation method based on improved YOLOv8n for ship equipment disassembly and assembly[J]. Chinese Journal of Ship Research, 2025, 20(2): 140

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Ship Intelligent O&M, and Fault Diagnosis

    Received: Apr. 28, 2024

    Accepted: --

    Published Online: May. 15, 2025

    The Author Email:

    DOI:10.19693/j.issn.1673-3185.03902

    Topics