Convolutional Neural Network Fire Smoke Detection Based on Target Region

Lujia Feng; Huiqin Wang; Ke Wang; Ying Lu; Jia Wang

doi:10.3788/LOP57.161004

Journals >Laser & Optoelectronics Progress >Volume 57 >Issue 16 >Page 161004 > Article

Laser & Optoelectronics Progress
Vol. 57, Issue 16, 161004 (2020)

Convolutional Neural Network Fire Smoke Detection Based on Target Region

Lujia Feng, Huiqin Wang^*, Ke Wang, Ying Lu, and Jia Wang

Author Affiliations

School of Information and Control Engineering, Xi'an University of Architecture and Technology, Xi'an, Shaanxi 710055, China

show less

DOI: 10.3788/LOP57.161004 Cite this Article Set citation alerts

Lujia Feng, Huiqin Wang, Ke Wang, Ying Lu, Jia Wang. Convolutional Neural Network Fire Smoke Detection Based on Target Region[J]. Laser & Optoelectronics Progress, 2020, 57(16): 161004 Copy Citation Text

show less

Fig. 1. Flowchart of background difference method

Download full size

Fig. 2. MD_CNN model structure diagram

Download full size

Fig. 3. Flow chart of MD_CNN model detection fire smoke

Download full size

Fig. 4. Smoke target region. (a) Original image; (b) smoke threshold image; (c) smoke target region

Download full size

Fig. 5. ReLU activation function

Download full size

Type	Feature map size	Kernel size	Stride
Conv1	320×240×32	5×5	1
Pool1	160×120×32	2×2	2
Conv2	160×120×64	5×5	1
Pool2	80×60×64	2×2	2
Conv3	80×60×128	3×3	1
Pool3	40×30×128	2×2	2
Conv4	40×30×128	3×3	1
Pool4	20×15×128	2×2	2
FC5	1024
FC6	1024
FC7	2

Table 1. Smoke detection layer parameters

Algorithm	ACC	TPR	FNR
CNN	89.57	94.28	10.43
MD_CNN	93.72	97.25	6.28

Table 2. Smoke recognition results of CNN and MD_CNN%

Model	ACC /%	TPR /%	FNR /%	Number of Layers
Model	ACC /%	TPR /%	FNR /%	Conv	Pooling	FC
Conv2_FC3	83.52	89.73	16.48	2	2	3
Conv3_FC3	88.43	93.68	11.57	3	3	3
Conv4_FC3	93.72	97.25	6.28	4	4	3
Conv5_FC3	92.93	95.89	7.07	5	5	3
Conv6_FC3	92.16	96.54	7.84	6	6	3
Conv7_FC3	90.04	94.39	9.96	7	7	3

Table 3. Comparison of structure and performance indicators of different models

Algorithm	ACC	TPR	FNR
MD_CNN	93.72	97.25	6.28
Algorithm 1	83.27	94.13	11.24
Algorithm 2	86.53	95.62	9.13
Algorithm 3	89.75	96.32	8.54

Table 4. Classification performance index values of four algorithms%

Lujia Feng, Huiqin Wang, Ke Wang, Ying Lu, Jia Wang. Convolutional Neural Network Fire Smoke Detection Based on Target Region[J]. Laser & Optoelectronics Progress, 2020, 57(16): 161004

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information