High-quality training data is essential for enhancing the robustness of object detection models. Within the maritime domain, obtaining a diverse real image dataset is particularly challenging due to the difficulty of capturing sea images with the presence of maritime objects , especially in stormy conditions. These challenges arise due to resource limitations, in addition to the unpredictable appearance of maritime objects. Nevertheless, acquiring data from stormy conditions is essential for training effective maritime detection models, particularly for search and rescue, where real-world conditions can be unpredictable. In this work, we introduce SafeSea, which is a stepping stone towards transforming actual sea images with various Sea State backgrounds while retaining maritime objects. Compared to existing generative methods such as Stable Diffusion Inpainting~\cite{stableDiffusion}, this approach reduces the time and effort required to create synthetic datasets for training maritime object detection models. The proposed method uses two automated filters to only pass generated images that meet the criteria. In particular, these filters will first classify the sea condition according to its Sea State level and then it will check whether the objects from the input image are still preserved. This method enabled the creation of the SafeSea dataset, offering diverse weather condition backgrounds to supplement the training of maritime models. Lastly, we observed that a maritime object detection model faced challenges in detecting objects in stormy sea backgrounds, emphasizing the impact of weather conditions on detection accuracy. The code, and dataset are available at https://github.com/martin-3240/SafeSea.
The Once-For-All (OFA) method offers an excellent pathway to deploy a trained neural network model into multiple target platforms by utilising the supernet-subnet architecture. Once trained, a subnet can be derived from the supernet (both architecture and trained weights) and deployed directly to the target platform with little to no retraining or fine-tuning. To train the subnet population, OFA uses a novel training method called Progressive Shrinking (PS) which is designed to limit the negative impact of interference during training. It is believed that higher interference during training results in lower subnet population accuracies. In this work we take a second look at this interference effect. Surprisingly, we find that interference mitigation strategies do not have a large impact on the overall subnet population performance. Instead, we find the subnet architecture selection bias during training to be a more important aspect. To show this, we propose a simple-yet-effective method called Random Subnet Sampling (RSS), which does not have mitigation on the interference effect. Despite no mitigation, RSS is able to produce a better performing subnet population than PS in four small-to-medium-sized datasets; suggesting that the interference effect does not play a pivotal role in these datasets. Due to its simplicity, RSS provides a 1.9× reduction in training times compared to PS. A 6.1× reduction can also be achieved with a reasonable drop in performance when the number of RSS training epochs are reduced. Code available at https://github.com/Jordan-HS/RSS-Interference-CVPRW2022
In 2019 SWIR Vision Systems introduced its 2.1 MP Acuros cameras to the industrial imaging market, becoming the first company globally to commercialize high resolution, quantum‐dot based image sensors. Since this product introduction, SWIR Vision Systems has continued to advance the performance of its colloidal quantum dot detector architecture. These advances include demonstrating detectors with 940 nm QE's > 50% and extended wavelength eSWIR detectors with spectral response from 350 nm to 2100 nm. This paper will provide an overview of our approach to fabricating focal plane arrays, will describe recent results fabricating Vis‐SWIR and eSWIR CQD® detector arrays, and will show imaging demonstrations of these sensors in a variety of applications.
This paper presents the design of a locomotion system for an unmanned ground vehicle to be used in minefield reconnaissance and mapping missions. The paper describes the analysis conducted to quantify the characteristics of rough terrain of the landmine contaminated area and its implications on selecting an efficient locomotion system. A comparative study based on 2-D and 3-D modeling is conducted between three 6-wheeled vehicles with articulated suspension. The optimal design is to be implemented within MineProbe project. MineProbe: A Distributed Mobile Sensor System for Minefield Reconnaissance and Mapping in Egypt is an applied research project that aims at developing a novel minefield reconnaissance and mapping system in Egypt focusing on North West Coast (NWC) of Egypt as location of the action.
It is well established that when steel components are cleaned by Ultra High Pressure Waterjetting (UHP-WJ) the surface begins to oxidize or ‘flash rust’ (FR) within a short period of time. FR has a major impact on subsequent coating application on these components and in most cases only a light FR surface is acceptable. Currently, there is no quantitative or semi-quantitative technique to characterize or categorize the level (or grade) of FR. However, descriptive and visual standards developed by SSPC and NACE are available. These standards are routinely used in the waterjetting industry but they are subjective in nature. Attempts have been made in the past or are presently being made by different entities to come up with a more definitive methodology but with limited success. The present paper discusses the application of electrochemical techniques for characterizing FR surfaces in a quantitative/semi-quantitative manner.
