Figure - available from: Machine Vision and Applications
This content is subject to copyright. Terms and conditions apply.
Qualitative comparison of results on Places2 validation dataset with gated convolution (GC), region normalization (RN), and the proposed method
Source publication
In this work, we propose a two-stage architecture to perform image inpainting from coarse to fine. The framework extracts advantages from different designs in the literature and integrates them into the inpainting network. We apply region normalization to generate coarse blur results with the correct structure. Then, contextual attention is applied...
Citations
... Recent work in the field of pattern recognition and machine learning [5,48] has provided significant opportunities for automatic extraction of information based on big data [28,49]. This is largely driven by the deep learning wave [21], which describes the most representative and discriminative features through hierarchical multilayer neural networks. ...
The accurate and efficient detection of shoppers and classification by age group, gender, and cart or trolley in supermarkets is essential for strategic retail planning. With the advent of deep learning algorithms, various models based on Convolutional Neural Networks (CNNs) have been proposed for object detection in high-resolution spatial images. In this article, we proposed an architecture that consists of two phases: a first phase of shopper detection where we conducted a comparative study of three well-established CNN-based models, namely Single Shot Multi-Box Detector (SSD), You Look Only Once-v8 (YOLOv8), and Faster R-CNN, to detect shoppers by age group, gender, and shopping basket. Transfer learning and fine-tuning approaches were implemented to train the models. The evaluation results for accuracy and efficiency show that YOLOv8 achieved the best performance in terms of mean Average Precision (mAP), Frames Per Second (FPS) metrics, and visual inspection. SSD demonstrated an advantage in terms of detection speed with an FPS twice as high as Faster R-CNN, although their mAP was close on the test set. The trained models were also applied to two independent test sets, proving their transferability and the importance of higher resolution images for accuracy improvement. In the second phase, we conducted a comparative study among the classification models Residual Network 50 (ResNet50), the Visual Geometry Group(VGG) 16 and 19 family, and Densely Connected Convolutional Network 121 (DenseNet121). We found a positive prediction rate of 99.28% for the VGG16 model and 98.55% for VGG19, with the others having a rate that is not too far off.
Droughts are among the most devastating natural hazards, occurring in all regions with different climate conditions. The impacts of droughts result in significant damages annually around the world. While drought is generally described as a slow‐developing hazardous event, a rapidly developing type of drought, the so‐called flash drought has been revealed by recent studies. The rapid onset and strong intensity of flash droughts require accurate real‐time monitoring. Addressing this issue, a Generative Adversarial Network (GAN) is developed in this study to monitor flash droughts over the Contiguous United States (CONUS). GAN contains two models: (a) discriminator and (b) generator. The developed architecture in this study employs a Markovian discriminator, which emphasizes the spatial dependencies, with a modified U‐Net generator, tuned for optimal performance. To determine the best loss function for the generator, four different networks are developed with different loss functions, including Mean Absolute Error (MAE), adversarial loss, a combination of adversarial loss with Mean Square Error (MSE), and a combination of adversarial loss with MAE. Utilizing daily datasets collected from NLDAS‐2 and Standardized Soil Moisture Index (SSI) maps, the network is trained for real‐time daily SSI monitoring. Comparative assessments reveal the proposed GAN's superior ability to replicate SSI values over U‐Net and Naïve models. Evaluation metrics further underscore that the developed GAN successfully identifies both fine‐ and coarse‐scale spatial drought patterns and abrupt changes in the SSI temporal patterns that is important for flash drought identification.
