Object and spatial discrimination makes weakly supervised local feature better.

Local feature extraction plays a crucial role in numerous critical visual tasks. However, there remains room for improvement in both descriptors and keypoints, particularly regarding the discriminative power of descriptors and the localization precision of keypoints. To address these challenges, this study introduces a novel local feature extraction pipeline named OSDFeat (Object and Spatial Discrimination Feature). OSDFeat employs a decoupling strategy, training descriptor and detection networks independently. Inspired by semantic correspondence, we propose an Object and Spatial Discrimination ResUNet (OSD-ResUNet). OSD-ResUNet captures features from the feature map that differentiate object appearance and spatial context, thus enhancing descriptor performance. To further improve the discriminative capability of descriptors, we propose a Discrimination Information Retained Normalization module (DIRN). DIRN complementarily integrates spatial-wise normalization and channel-wise normalization, yielding descriptors that are more distinguishable and informative. In the detection network, we propose a Cross Saliency Pooling module (CSP). CSP employs a cross-shaped kernel to aggregate long-range context in both vertical and horizontal dimensions. By enhancing the saliency of keypoints, CSP enables the detection network to effectively utilize descriptor information and achieve more precise localization of keypoints. Compared to the previous best local feature extraction methods, OSDFeat achieves Mean Matching Accuracy of 79.4% in local feature matching task, improving by 1.9% and achieving state-of-the-art results. Additionally, OSDFeat achieves competitive results in Visual Localization and 3D Reconstruction. The results of this study indicate that object and spatial discrimination can improve the accuracy and robustness of local feature, even in challenging environments. The code is available at https://github.com/pandaandyy/OSDFeat.

Wireless Flexible System for Highly Sensitive Ammonia Detection Based on Polyaniline/Carbon Nanotubes.

Ammonia (NH3) is a harmful atmospheric pollutant and an important indicator of environment, health, and food safety conditions. Wearable devices with flexible gas sensors offer convenient real-time NH3 monitoring capabilities. A flexible ammonia gas sensing system to support the internet of things (IoT) is proposed. The flexible gas sensor in this system utilizes polyaniline (PANI) with multiwall carbon nanotubes (MWCNTs) decoration as a sensitive material, coated on a silver interdigital electrode on a polyethylene terephthalate (PET) substrate. Gas sensors are combined with other electronic components to form a flexible electronic system. The IoT functionality of the system comes from a microcontroller with Wi-Fi capability. The flexible gas sensor demonstrates commendable sensitivity, selectivity, humidity resistance, and long lifespan. The experimental data procured from the sensor reveal a remarkably low detection threshold of 0.3 ppm, aligning well with the required specifications for monitoring ammonia concentrations in exhaled breath gas, which typically range from 0.425 to 1.8 ppm. Furthermore, the sensor demonstrates a negligible reaction to the presence of interfering gases, such as ethanol, acetone, and methanol, thereby ensuring high selectivity for ammonia detection. In addition to these attributes, the sensor maintains consistent stability across a range of environmental conditions, including varying humidity levels, repeated bending cycles, and diverse angles of orientation. A portable, stable, and effective flexible IoT system solution for real-time ammonia sensing is demonstrated by collecting data at the edge end, processing the data in the cloud, and displaying the data at the user end.

Effects of Carbon Nanofibers and Carboxyl Functionalized Multi-Walled Carbon Nanotubes on Mechanical Damping Behavior of Cement Paste.

Different ratios of carbon nanofibers (CNF) and carboxyl functionalized multi-walled carbon nanotubes (CMWCNT), dispersed by the polycarboxylate based superplasticizer, were added in this study to investigate mechanical damping behavior of cement paste. The additions of CNF and CMWCNT simultaneously improved loss tangent, storage modulus and loss modulus of cement paste to different extents at 0.2 Hz the frequency. The cement paste exhibited higher storage modulus and loss modulus with addition of CNF. The addition of CMWCNT provided higher loss tangent but lower modulus compared to the CNF modified cement paste. The damping mechanisms for cement incorporated with CNF and CMWCNT were explained by the effective "stick-slip" motion which was generated by the possible sliding among the nanofilaments at the interface between the fibers and cement particles. The enhanced stiffness was attributed to the optimized microstructure and bridge effect induced by the nanoscale nanofilaments. Both total porosity and mesoporosity decreased due to the nano-filler effect from the carbon nanofilaments. The SEM technique indicated that excessive addition of CNF/CMWCNT should be avoided to minimize possible fiber agglomeration in the cement matrix.