基于YOLO-RAMS的计算机随机存取存储器插槽旋转检测算法

doi:10.3778/j.issn.1002-8331.2502-0107

摘要/Abstract

摘要： 针对计算机随机存取存储器智能化安装场景中，需要快速精确定位插槽和计算其角度等问题，提出一种改进YOLOv8n-obb的计算机随机存取存储器插槽旋转检测算法YOLO-RAMS。在主干高层设计扩张重参数化残差模块，增强网络捕获稀疏模式的能力，充分提取更丰富的语义特征，并构建多速率扩张卷积金字塔模块，提高模型对全局上下文和细节信息的关注度；在颈部设计双重维度感知特征融合扩散网络，专注于对不同维度特征的自适应选择和精细融合，以提升多尺度目标的显著性；在头部设计特征交互动态检测头并添加P2层，增加头部对交互特征的学习以及增强头部的动态特性和小目标的显著性，进一步提高检测精度；引入瓶颈注意力模块，突出关键信息，强化模型表征能力。实验结果表明，YOLO-RAMS的准确率、召回率、mAP@0.5和mAP@0.5：0.95达到89.2%、78.2%、90.1%和57.4%，相比原模型分别提高6.8、4.4、5.7和6.6个百分点，平均角度误差1.7°，参数量为2.69×106，检测帧率达到172.2 FPS，该算法有效减少了误检、漏检及角度误差，具有较优的实际应用性能。

关键词: 随机存取存储器, 旋转检测, YOLOv8n-obb, 扩张重参数化残差, 多速率扩张, 双重维度感知, 特征交互

Abstract: Aiming at the problems of quickly and accurately locating slots and calculating their angles in the intelligent installation scenario of computer random access memory, an improved YOLOv8n-obb computer random access memory slot rotation detection algorithm YOLO-RAMS is proposed. Firstly, the dilated re-parameterized residual module is designed at the high level of the backbone to enhance the ability of the network to capture sparse patterns and fully extract richer semantic features, and the multi-rate dilated convolution pyramid module is constructed to improve the model’s focus on global context and detail information. Secondly, a dual dimensional-aware feature fusion diffusion network is designed in the neck, focusing on adaptive selection and fine fusion of features of different dimensions to enhance the saliency of multiscale targets; and a feature interaction dynamic detection head is designed in the head and added with a P2 layer, increasing the head’s learning of interacting features and enhancing the head’s dynamic characteristics and the saliency of small targets to further improve the detection accuracy. Finally, the bottleneck attention module is introduced to highlight key information and strengthen the model representation capability. The experimental results show that the accuracy, recall, mAP@0.5 and mAP@0.5：0.95 of YOLO-RAMS reach 89.2%, 78.2%, 90.1% and 57.4%, which are 6.8, 4.4, 5.7 and 6.6 percentage points higher than the original model, the average angular error is 1.7°, the number of parameters is 2.69×106, and the detection frame rate reaches 172.2 FPS. The proposed algorithm effectively reduces the false detection, missed detection and angular error, and has superior performance in practical applications.

Key words: random access memory, rotation detection, YOLOv8n-obb, dilated re-parameterized residual, multi-rate dilated, dual dimension-aware, feature interaction

陈奥, 王琨, 贺昊辰. 基于YOLO-RAMS的计算机随机存取存储器插槽旋转检测算法[J]. 计算机工程与应用, 2025, 61(17): 147-158.

CHEN Ao, WANG Kun, HE Haochen. YOLO-RAMS-Based Algorithm for Rotational Detection of Computer Random Access Memory Slot[J]. Computer Engineering and Applications, 2025, 61(17): 147-158.

参考文献

[1] ARABIAN J. Computer integrated electronics manufacturing and testing[M]. Boca Raton: CRC Press, 2020: 1-7.
[2] CHIEW J H, SUNG A N. Augmented reality application for laptop assembly with assembly complexity study[J]. The International Journal of Advanced Manufacturing Technology, 2022, 120(1): 1149-1167.
[3] 甘小红, 覃志东, 蔡勇, 等. 一种基于SIFT的改进优化特征匹配算法[J]. 智能计算机与应用, 2021, 11(11): 5-9.
GAN X H, QIN Z D, CAI Y, et al. An optimized image feature matching algorithm based on SIFT[J]. Intelligent Computer and Applications, 2021, 11(11): 5-9.
[4] SU N, WANG J, XU G, et al. Infrared visual sensing detection of groove width for swing arc narrow gap welding[J]. Sensors (Basel), 2022, 22(7): 2555.
[5] NAJI O A A M, SHAH H N M, ANWAR N S N, et al. Square groove detection based on F?rstner with Canny edge operator using laser vision sensor[J]. The International Journal of Advanced Manufacturing Technology, 2023, 125(5): 2885-2894.
[6] 张放. 基于并联机器人组合的内存条装配技术研究[D]. 上海: 上海交通大学, 2022.
ZHANG F. Research on memory module assembly technology based on parallel robots combination[D]. Shanghai: Shanghai Jiao Tong University, 2022.
[7] WANG Y, WU Z H, DUAN X Y, et al. An micro head positio-ning slot recognition detection system based on machine vision[J]. IOP Conference Series: Materials Science and Engineering, 2019, 677(4): 042075.
[8] 王丽威, 马丽艳, 李功燕. 基于直线段检测的SIM卡槽图像中切割线定位算法[J]. 数据采集与处理, 2018, 33(1): 161-170.
WANG L W, MA L Y, LI G Y. Cutting line detection for SIM card slot images based on line segment detection[J]. Journal of Data Acquisition and Processing, 2018, 33(1): 161-170.
[9] 陈俊英, 李朝阳, 黄汉涛, 等. 并行特征提取和渐进特征融合的计算机主板装配缺陷检测[J]. 光学精密工程, 2024, 32(10): 1622-1637.
CHEN J Y, LI Z Y, HUANG H T, et al. Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion[J]. Optics and Precision Engineering, 2024, 32(10): 1622-1637.
[10] CHANG B R, TSAI H F, CHANG F Y. Applying advanced lightweight architecture DSGSE-YOLOv5 to rapid chip contour detection[J]. Electronics, 2024, 13(1): 10.
[11] CHEN Y C, HWANG R H, CHEN M Y, et al. Screw slot quality inspection system based on tactile network[J]. ACM Transactions on Internet Technology, 2021, 21(4): 1-17.
[12] 熊沛霖, 陈雯柏, 张波, 等. 面向异形零件装配的精准视觉检测和分割算法[J]. 兵器装备工程学报, 2024, 45(5): 239-247.
XIONG P L, CHEN W B, ZHANG B, et al. Accurate visual detection and segmentation algorithm for special-shaped parts assembly[J]. Journal of Ordnance Equipment Engineering, 2024, 45(5): 239-247.
[13] TU G M, QIN J H, XIONG N. Algorithm of computer mainboard quality detection for real-time based on QD-YOLO[J]. Electronics, 2022, 11(15): 2424.
[14] ZHANG Z H, ZHOU Q Y, JIN H P, et al. A real-time intelligent valve monitoring approach through cameras based on computer vision methods[J]. Sensors, 2024, 24(16): 5337.
[15] LIU Y, ZHOU T, XU J Y, et al. Rotating target detection method of concrete bridge crack based on YOLOv5[J]. Applied Sciences, 2023, 13(20): 11118.
[16] FANG Z, WANG X Y, ZHANG L, et al. YOLO-RSA: a multiscale ship detection algorithm based on optical remote sensing image[J]. Journal of Marine Science and Engineering, 2024, 12(4): 603.
[17] WEI H, LIU X, XU S, et al. DWRSeg: rethinking efficient acquisition of multi-scale contextual information for real-time semantic segmentation[J]. arXiv:2212.01173, 2022.
[18] DING X H, ZHANG Y Y, GE Y X, et al. UniRepLKNet: a universal perception large-kernel ConvNet for audio, video, point cloud, time-series and image recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2024: 5513-5524.
[19] XU S, ZHENG S C, XU W, et al. HCF-Net: hierarchical context fusion network for infrared small object detection[J]. arXiv:2403.10778, 2024.
[20] TIAN Z, SHEN C H, CHEN H, et al. FCOS: a simple and strong anchor-free object detector[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2020: 9626-9635.
[21] FENG C J, ZHONG Y J, GAO Y, et al. TOOD: task-aligned one-stage object detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2021: 3490-3499.
[22] XIONG Y W, LI Z Q, CHEN Y T, et al. Efficient deformable ConvNets: rethinking dynamic and sparse operator for vision applications[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2024: 5652-5661.
[23] ZHAO Y A, LYU W Y, XU S L, et al. DETRs beat YOLOs on real-time object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2024: 16965-16974.
[24] YANG J, LI C, DAI X, et al. Focal modulation networks[C]//Advances in Neural Information Processing Systems, 2022: 4203-4217.
[25] 李明, 何志奇, 党青霞, 等. 面向户外导盲场景的道路目标检测算法[J]. 计算机工程与应用, 2025, 61(9): 242-254.
LI M, HE Z Q, DANG Q X, et al. Road object detection algorithm for outdoor blind navigation scenarios[J]. Computer Engineering and Applications, 2025, 61(9): 242-254.
[26] DAI J F, QI H Z, XIONG Y W, et al. Deformable convolutional networks[C]//Proceedings of the IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 764-773.
[27] ZHU X Z, HU H, LIN S, et al. Deformable ConvNets V2: more deformable, better results[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 9300-9308.
[28] LI H H, ZHANG Y W, ZHANG Y, et al. DCNv3: towards next generation deep cross network for CTR prediction[J]. arXiv:2407.13349, 2024.
[29] LI W T, CHEN Y J, HU K X, et al. Oriented RepPoints for aerial object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 1819-1828.
[30] HAN J M, DING J, XUE N, et al. ReDet: a rotation-equivariant detector for aerial object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 2785-2794.
[31] XU Y, FU M, WANG Q, et al. Gliding vertex on the horizontal bounding box for multi-oriented object detection[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2021, 43(4): 1452-1459.
[32] DING J, XUE N, LONG Y, et al. Learning RoI transformer for oriented object detection in aerial images[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 2844-2853.
[33] HAN J M, DING J, LI J, et al. Align deep features for oriented object detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-11.
[34] YANG X, YAN J C, FENG Z M, et al. R3Det: refined single-stage detector with feature refinement for rotating object[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2021: 3163-3171.