An improved maritime object detection algorithm, SRC-YOLO, based on the YOLOv4-tiny, is proposed in the foggy environment to address the issues of false detection, missed detection, and low detection accuracy in complicated situations. To confirm the model’s validity, an ocean dataset containing various concentrations of haze, target angles, and sizes was produced for the research. Firstly, the Single Scale Retinex (SSR) algorithm was applied to preprocess the dataset to reduce the interference of the complex scenes on the ocean. Secondly, in order to increase the model’s receptive field, we employed a modified Receptive Field Block (RFB) module in place of the standard convolution in the Neck part of the model. Finally, the Convolutional Block Attention Module (CBAM), which integrates channel and spatial information, was introduced to raise detection performance by expanding the network model’s attention to the context information in the feature map and the object location points. The experimental results demonstrate that the improved SRC-YOLO model effectively detects marine targets in foggy scenes by increasing the mean Average Precision (mAP) of detection results from 79.56% to 86.15%.

To improve the performance of the forest fire smoke detection model and achieve a better balance between detection accuracy and speed, an improved YOLOv4 detection model (MoAm-YOLOv4) that combines a lightweight network and attention mechanism was proposed. Based on the YOLOv4 algorithm, the backbone network CSPDarknet53 was replaced with a lightweight network MobilenetV1 to reduce the model's size. An attention mechanism was added to the three channels before the output to increase its ability to extract forest fire smoke effectively. The algorithm used the K-means clustering algorithm to cluster the smoke dataset, and obtained candidate frames that were close to the smoke images; the dataset was expanded to 2000 images by the random flip expansion method to avoid overfitting in training. The experimental results show that the improved YOLOv4 algorithm has excellent detection effect. Its mAP can reach 93.45%, precision can get 93.28%, and the model size is only 45.58 MB. Compared with YOLOv4 algorithm, MoAm-YOLOv4 improves the accuracy by 1.3% and reduces the model size by 80% while sacrificing only 0.27% mAP, showing reasonable practicability.