Coaxial Tracking Method of Infrared Search and Tracking System Based on Photoelectric Calibration

Zhang, Feng

doi:10.1007/978-981-19-6613-2_368

Cited by 1 publication

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The full name of the MSIA module is the multistage information aggregation module. Some studies have pointed out that the input I is convolved with K (1) first, convolved with input I and K (2) , and then added, and the convolved result is the same as that of K (1) and K (2) . Then, it is added point by point to I, as shown in Equation (1).…”

Section: Msia Modulementioning

confidence: 99%

“…Some studies have pointed out that the input I is convolved with K (1) first, convolved with input I and K (2) , and then added, and the convolved result is the same as that of K (1) and K (2) . Then, it is added point by point to I, as shown in Equation (1). For AC-Net [27], the author verified the importance of the skeleton in the square convolution kernel d×d and divided the d×d convolution into three-way d×d, 1×d, and d×1 convolution.…”

Section: Msia Modulementioning

confidence: 99%

“…Each captures the remote correlation of the input feature map along each spatial direction. Compared with the common SE and CBAM modules, this model has the following advantages: (1) The orientation-related position information is integrated into the channel so that the model has a stronger ability to locate and identify the target; (2) it is a lightweight module that can be easily plugged into the network. In the field of target detection, shallow features contain accurate location information of targets and have better recognition abilities for small targets, while deep features contain the semantic information of targets and are often used to detect medium and large targets.…”

Section: Bottleneck Network Vs Prediction Networkmentioning

confidence: 99%

“…Infrared target detection models are widely used in some fields, such as in assisted automobile driving and shipborne infrared search [1]. Traditional methods include the following three categories: (1) threshold segmentation, (2) template matching, and (3) frame difference.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MSIA-Net: A Lightweight Infrared Target Detection Network with Efficient Information Fusion

Zhou

et al. 2023

Entropy

View full text Add to dashboard Cite

In order to solve the problems of infrared target detection (i.e., the large models and numerous parameters), a lightweight detection network, MSIA-Net, is proposed. Firstly, a feature extraction module named MSIA, which is based on asymmetric convolution, is proposed, and it can greatly reduce the number of parameters and improve the detection performance by reusing information. In addition, we propose a down-sampling module named DPP to reduce the information loss caused by pooling down-sampling. Finally, we propose a feature fusion structure named LIR-FPN that can shorten the information transmission path and effectively reduce the noise in the process of feature fusion. In order to improve the ability of the network to focus on the target, we introduce coordinate attention (CA) into the LIR-FPN; this integrates the location information of the target into the channel so as to obtain more expressive feature information. Finally, a comparative experiment with other SOTA methods was completed on the FLIR on-board infrared image dataset, which proved the powerful detection performance of MSIA-Net.

show abstract