Edge feature and detail-aware network integrated YOLOv8s algorithm for hip joint keypoint detection

Jia Lv; Xunlu Duan; Xin Chen

doi:10.12086/oee.2025.240281

Opto-Electronic Engineering, Volume. 52, Issue 3, 240281(2025)

Edge feature and detail-aware network integrated YOLOv8s algorithm for hip joint keypoint detection

Jia Lv^1,2, Xunlu Duan¹, and Xin Chen^3、*

Author Affiliations

¹College of Computer and Information Sciences, Chongqing Normal University, Chongqing 401331, China

²National Center for Applied Mathematics in Chongqing, Chongqing Normal University, Chongqing 401331, China

³Department of Radiology, Children’s Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Pediatrics, Chongqing 400014, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(17)

Fig. 1. Diagnostic reference for developmental dysplasia of the hip^[8]

Download full size

View in Article

Fig. 2. Overall network structure diagram of EDA-YOLOv8s

Download full size

View in Article

Fig. 3. Structural diagram of EFEM module

Download full size

View in Article

Fig. 4. Comparison diagram of three neck network structures in YOLOv8s. (a) Neck; (b) Small-neck; (c) DAN

Download full size

View in Article

Fig. 5. Structural diagram of DSFM module

Download full size

View in Article

Fig. 6. Structure diagram of CSP-OKM module and associated submodules. (a) CSP-OKM; (b) FSAM; (c) DCAM; (d) OKM

Download full size

View in Article

Fig. 7. Heatmaps of hip X-ray image after different improved modules. (a) YOLOv8s; (b) YOLOv8s+EFEM; (c) YOLOv8s+DAN; (d) EDA-YOLOv8s

Download full size

View in Article

Fig. 8. The relationship diagram between APE and AAE after using different improvement modules

Download full size

View in Article

Fig. 9. Experimental results of successful detection rate of keypoint

Download full size

View in Article

Fig. 10. Comparison of keypoint localization error distribution. (a) YOLOv8s; (b) EDA-YOLOv8s

Download full size

View in Article

Fig. 11. Visualization comparison of detection results between EDA-YOLOv8s and YOLOv8s

Download full size

View in Article

Fig. 12. Visualization comparison of keypoint angle between EDA-YOLOv8s and YOLOv8s. (a) YOLOv8s; (b) EDA-YOLOv8s

Download full size

View in Article

Table 1. Evaluation index and computational formula
View table
View in Article
Table 1. Evaluation index and computational formula
Index Calculation formula
PEL_l $(\sum_{i = 1}^{n} \sqrt{T_{l i} - P_{l i}}) / n$
APE $(P E L_{RASM} + P E L_{RTCC} + P E L_{LTCC} + P E L_{LASM}) / 4$
EA_k $(\sum_{i = 1}^{n} | T_{k i} - P_{k i} |) / n$
AAE $(E A_{R} + E A_{L}) / 2$
SDR ${i : \sqrt{{(x_{T_{l i}} - x_{P_{l i}})}^{2} + {(y_{T_{l i}} - y_{P_{l i}})}^{2}} ⩽ z} \times 100 % / n$
MDR $\frac{m}{n} \times 100 %$

Table 2. The ablation and edge detection operator comparison experiment results in the EFEM module
View table
View in Article
Table 2. The ablation and edge detection operator comparison experiment results in the EFEM module
EFEM (edge detection operator) BN APE/pixel AAE/(°)
× × 4.5150 1.6504
Sobel × 4.3652 1.5842
Sobel √ 4.3459 1.5565
Prewitt √ 4.3470 1.5555
Laplacian √ 4.4130 1.6934

Table 3. Comparison experiment results of three neck network algorithms
View table
View in Article
Table 3. Comparison experiment results of three neck network algorithms
Neck network APE/pixel AAE/(°) FLOPs/G
YOLOv8s (neck) 4.5150 1.6504 28.8
YOLOv8s+ (small-neck) 4.7150 1.7322 36.9
YOLOv8s+ (DAN) 4.3281 1.5451 34.8

Table 4. The influence of different improved modules on the algorithm

View table

View in Article

Table 4. The influence of different improved modules on the algorithm

Detection algorithm	PEL_RASM/pixel	PEL_RTCC/pixel	PEL_LTCC/pixel	PEL_LASM/pixel	APE/pixel	EA_R/(°)	EA_L/(°)	AAE/(°)
YOLOV8s	4.8476	3.7404	3.4674	6.0045	4.5150	1.6250	1.6757	1.6504
YOLOv8s+EFEM	4.6133	3.7692	3.2258	5.7752	4.3459	1.4886	1.6244	1.5565
YOLOv8s+DAN	4.1284	3.6947	3.3784	6.1110	4.3281	1.2587	1.8315	1.5451
EDA-YOLOv8s	4.2851	3.4870	3.2530	5.8107	4.2090	1.3420	1.6324	1.4872

Table 5. Comparison experiment results of different keypoint detection algorithms

View table

View in Article

Table 5. Comparison experiment results of different keypoint detection algorithms

Detection algorithm	PEL_RASM/pixel	PEL_RTCC/pixel	PEL_LTCC/pixel	PEL_LASM/pixel	APE/pixel	EA_R/(°)	EA_L/(°)	AAE/(°)	MDR/%
YOLOv8n	5.5012	3.8100	3.2056	5.9099	4.6067	1.4698	1.7188	1.5943	0
YOLOv8s	4.8476	3.7404	3.4674	6.0045	4.5150	1.6250	1.6757	1.6504	0
YOLOv9s	4.6463	3.7893	3.4716	6.2252	4.5331	1.4878	1.8033	1.6456	0
YOLOv10s	4.6806	3.7820	3.3007	5.9486	4.4280	1.4594	1.8313	1.6454	6.0
PN-UNet	4.5411	3.7115	3.6881	6.2045	4.5363	1.4384	1.6533	1.5458	0
CircleNet	4.8348	3.7234	3.6113	5.8375	4.5017	1.4988	1.7996	1.6492	0
CBA-YOLOv5s	4.3828	3.6464	3.5506	6.3773	4.4893	1.3235	1.8488	1.5861	0
EDA-YOLOv8s	4.2851	3.4870	3.2530	5.8107	4.2090	1.3420	1.6324	1.4872	0

Tools

Get Citation

Copy Citation Text

Jia Lv, Xunlu Duan, Xin Chen. Edge feature and detail-aware network integrated YOLOv8s algorithm for hip joint keypoint detection[J]. Opto-Electronic Engineering, 2025, 52(3): 240281

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites