J-MSF：A new infrared dim and small target detection algorithm based on multi-channel and multiscale

Guogang Wang; Zhaojin Sun; Yunpeng Liu

doi:10.3788/IRLA20210459

Journals >Infrared and Laser Engineering >Volume 51 >Issue 5 >Page 20210459 > Article

Infrared and Laser Engineering
Vol. 51, Issue 5, 20210459 (2022)

J-MSF：A new infrared dim and small target detection algorithm based on multi-channel and multiscale

Guogang Wang¹, Zhaojin Sun¹, and Yunpeng Liu²

Author Affiliations

¹College of Information Engineering, Shenyang University of Chemical Technology, Shenyang 110142, China

²Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China

show less

DOI: 10.3788/IRLA20210459 Cite this Article

Guogang Wang, Zhaojin Sun, Yunpeng Liu. J-MSF：A new infrared dim and small target detection algorithm based on multi-channel and multiscale[J]. Infrared and Laser Engineering, 2022, 51(5): 20210459 Copy Citation Text

show less

Fig. 1. YOlOv3 flow chart of detection

Download full size | View in the Article

Fig. 2. Detection flow chart of J-MSF

Download full size | View in the Article

Fig. 3. Structural unit of JAnet

Download full size | View in the Article

Fig. 4. Network structure of J-MSF

Download full size | View in the Article

Fig. 5. Pixel value of target in test set sequence

Download full size | View in the Article

Fig. 6. Training loss curve

Download full size | View in the Article

Fig. 7. (a) Mark contrast box; (b) YOLO-Tiny detection result; (c) YOLOv3 detection result; (d) YOLOv3+SPP detection result; (e) Gaussian YOLOv3+SPP detection result; (f) YOLOv4 detection result; (g) J-MSF detection result

Download full size | View in the Article

Fig. 8. (a) YOLO-Tiny Precision-R curve; (b) YOLOv3 Precision-R curve; (c) YOLOv3+SPP Precision-R curve; (d) Gaussian YOLOv3+SPP Precision-R curve; (e) YOLOv4 Precision-R curve; (f) J-MSF Precision-R curve

Download full size | View in the Article

Fig. 9. Mainstream algorithm FPS-AP curve

Download full size | View in the Article

Fusion map/layer	Kernel size	Output size	Stride	Channel
Basic-feature map	-	8×8	-	1024
Artery-feature map	-	16×16	-	768
Detection map 1	-	32×32	-	30
Detection map 2	-	64×64	-	30
Detection map 3	-	128×128	-	30
Maxpooling 1	3	32×32	1	128
Maxpooling 2	5	32×32	1	128
Maxpooling 3	7	32×32	1	128

Table 1. Dimensions of network parameters

View in the Article

SNR region	3.26-3	3-2	2-1	1-0	0-(−1.97)	−3-(−20)
Data4	0	5	209	379	204	2
Data8	2	39	108	94	101	55
Data12	5	84	407	424	341	238
Data16	5	247	214	15	1	12
Data20	0	12	155	197	29	8
Total	12	387	1093	1109	676	315

Table 2. SNR data distribution table of test set

View in the Article

Model	$\mathop X\nolimits_{FN} $	R	AP
Darknet-53	458	87.2%	86.38%
Darknet-53-JA	343	90.0%	88.43%
J-MSF	217	94.0%	93.13%

Table 3. Contrast experiment of JAnet network

View in the Article

Darknet53	J-MSF	Loss	Fusion	Precision	R	AP	FPS
√	-	D	-	86%	87.20%	86.38%	66.3
√	-	D	√	92%	92.20%	92.74%	57.5
√	-	M	-	89%	94.04%	93.88%	71.9
√	-	M	√	82%	95.00%	93.47%	71.6
-	√	D	-	90%	94.00%	93.13%	59.0
-	√	D	√	90%	94.10%	93.46%	73.4
-	√	M	-	86%	95.85%	94.80%	66.8
-	√	M	√	88%	96.27%	96.29%	67.6

Table 4. Ablation study

View in the Article

Detection algorithm	$\mathop X\nolimits_{TP} $	$\mathop X\nolimits_{FP} $	$\mathop X\nolimits_{FN} $	Precision	R	AP
YOLO-Tiny	2389	1355	1203	59%	64%	45.32%
YOLOv3	3309	435	283	88%	92%	86.38%
YOLOv3+SPP[17]	3318	283	274	92%	92%	92.74%
Gaussian YOLOv3[18]+SPP	3407	758	185	78%	95%	93.60%
YOLOv4	3397	446	195	88%	95%	93.13%
J-MSF	3443	451	149	88%	96%	96.29%

Table 5. Results of infrared target detection by YOLO serial model

View in the Article

Detection algorithm	AP	FPS
Faster R-CNN^[20]	43.7%	35.2
SSD300^[21]	52.3%	154.7
RefineDet^[22]	63.9%	70.1
RetinaNet^[23]	65.4%	80.3
YOLOv3	86.4%	66.3
YOLOv4	93.1%	66.8
J-MSF	96.3%	67.6

Table 6. Comparison of mainstream algorithms

Guogang Wang, Zhaojin Sun, Yunpeng Liu. J-MSF：A new infrared dim and small target detection algorithm based on multi-channel and multiscale[J]. Infrared and Laser Engineering, 2022, 51(5): 20210459

Download Citation

Tools

Save the article for my favorites

Paper Information