Timely crack detection and evaluation is essential to prevent the further deterioration of the structural damage, and convolutional neural networks (CNNs) have potential applications due to their ...powerful feature extraction capability. Therefore, this paper employs some well-known CNN models to detect cracks and reveal the extraction features of the CNN. Subsequently, a state-of-the-art pixel-level segmentation CNN (DeepLab_v3+) was employed for crack segmentation, and the physical properties (the length and width, and so on) of the cracks were calculated according to the segmentation results. The results confirm that the CNN can effectively extract the features of different categories of cracks, and the transferred SqueezeNet has the best performance in the classification of the cracks, with competitive computational speed and accuracy, and the CNN trained on concrete cracks can also effectively detect cracks in asphalt pavement and walls. DeepLab_v3+ can also achieve the ideal performance of cracks segmentation (the MIoU, accuracy and F-score were 0.80, 97.5 and 0.78%, respectively). Finally, the proposed quantization method based on MATLAB platform achieves the desired accuracy, and the average relative errors of the length, average width, maximum width, area and cracking ratio were 7, 2, 22, 27 and 27%, respectively.
•A portable front face fluorescent system for detection of Al3+ was developed.•The PFFFS can be used to detect Al3+ in flour food within 10 min.•The PFFFS has the on-site and portable ...characteristics.
A simple and rapid method for in situ detection of aluminum in flour food was developed by using a self-made portable front face fluorescent system (PFFFS). The effects of pH, temperature, reaction time, protective agent and masking agent on the detection of Al3+ were investigated. The use of fluorescent probe protective agent, interfering ion masking agent, multi-point collection measurements and the working curves based on the analyte content in the real samples makes the present method have high accuracy, selectivity and reliability for in situ detection of Al3+ in flour foods. By comparison with the ICP-MS the accuracy and reliability of the present method were verified. The results showed that when 97 real samples were analyzed the Al3+ content values obtained by the present method and those obtained by ICP-MS method reached a highly significant correlation, with r ranging from 0.9747 to 0.9844. The self-made PFFFS combined with fluorescent probe does not require sample digestion, and can quickly detect Al3+ in flour food within 10 min. Therefore, the present method based on using FFFS has good practical application value for in-situ rapid detection of Al3+ in flour foods.
This paper presents a novel approach to substantially improve the detection accuracy of structural damage via a one-dimensional convolutional neural network (1-D CNN) and a decision-level fusion ...strategy. As structural damage usually induces changes in the dynamic responses of a structure, a CNN can effectively extract structural damage information from the vibration signals and classify them into the corresponding damage categories. However, it is difficult to build a large-scale sensor system in practical engineering; the collected vibration signals are usually non-synchronous and contain incomplete structure information, resulting in some evident errors in the decision stage of the CNN. In this study, the acceleration signals of multiple acquisition points were obtained, and the signals of each acquisition point were used to train a 1-D CNN, and their performances were evaluated by using the corresponding testing samples. Subsequently, the prediction results of all CNNs were fused (decision-level fusion) to obtain the integrated detection results. This method was validated using both numerical and experimental models and compared with a control experiment (data-level fusion) in which all the acceleration signals were used to train a CNN. The results confirmed that: by fusing the prediction results of multiple CNN models, the detection accuracy was significantly improved; for the numerical and experimental models, the detection accuracy was 10% and 16–30%, respectively, higher than that of the control experiment. It was demonstrated that: training a CNN using the acceleration signals of each acquisition point and making its own decision (the CNN output) and then fusing these decisions could effectively improve the accuracy of damage detection of the CNN.
Based on the classification ability of a convolutional neural network (CNN), this paper proposes a structural damage detection method in which a CNN is used to classify the location and level of ...damage in a structure. The dynamic responses are combined with the modal parameters of the structure as the inputs to the CNN to detect the damage. As structure damage can cause changes in multiple damage indicators, an individual indicator may not be enough to detect all damage scenarios. The combination of multiple damage indicators will provide more comprehensive information for damage situation. It is expected that this combination will overcome disadvantages of the damage index based on a single modal parameter. The finite element method was used to provide the training samples for the network. Damage in an element was introduced by reducing its Young’s modulus. Two cases were considered for the input of the CNN: the first used the modal strain energy only, and the second used the combination of modal strain energy and dynamic response (acceleration). The comparison results show that the inclusion of dynamic responses in the damage index significantly improves the correctness rate of structural damage detection and enhance the convergence of the network.
This paper compares the crack detection performance (in terms of precision and computational cost) of the YOLO_v2 using 11 feature extractors, which provides a base for realizing fast and accurate ...crack detection on concrete structures. Cracks on concrete structures are an important indicator for assessing their durability and safety, and real-time crack detection is an essential task in structural maintenance. The object detection algorithm, especially the YOLO series network, has significant potential in crack detection, while the feature extractor is the most important component of the YOLO_v2. Hence, this paper employs 11 well-known CNN models as the feature extractor of the YOLO_v2 for crack detection. The results confirm that a different feature extractor model of the YOLO_v2 network leads to a different detection result, among which the AP value is 0.89, 0, and 0 for ‘resnet18’, ‘alexnet’, and ‘vgg16’, respectively meanwhile, the ‘googlenet’ (AP = 0.84) and ‘mobilenetv2’ (AP = 0.87) also demonstrate comparable AP values. In terms of computing speed, the ‘alexnet’ takes the least computational time, the ‘squeezenet’ and ‘resnet18’ are ranked second and third respectively; therefore, the ‘resnet18’ is the best feature extractor model in terms of precision and computational cost. Additionally, through the parametric study (influence on detection results of the training epoch, feature extraction layer, and testing image size), the associated parameters indeed have an impact on the detection results. It is demonstrated that: excellent crack detection results can be achieved by the YOLO_v2 detector, in which an appropriate feature extractor model, training epoch, feature extraction layer, and testing image size play an important role.
•Sitagliptin is a dipeptidyl peptidase 4 inhibitor used as an anti-diabetic drug.•We investigate the protective effect of sitagliptin on the cognition in AD.•Sitagliptin treatment effectively ...protected the cognition function of the AD mice.
Sitagliptin is an anti-diabetic drug and its effects on Alzheimer’s disease (AD) remain controversial. This study aimed to investigate the protective effect of sitagliptin on the cognition in AD and its underlying molecular mechanism.
The APP/PS1 (a model of AD) mice received daily gastric gavage administration of sitagliptin (20 mg/kg) for 8 weeks. Then animals were subjected to behavioral experiment or sacrificed to histological staining and protein level analysis.
The MWM test showed that sitagliptin treatment significantly reduced the escape latency times in APP/PS1 mice in the learning phase (day 3–5) and elongated the time spent in the target quadrant in the probe test. Sitagliptin significantly reduced amyloid plaque deposition and elevated the spine density and the protein levels of synaptoneurosome GluA1- and GluA2-containing AMPA receptor (GluA1R and GluA2R) in the brain of the APP/PS1 mice. Sitagliptin treatment significantly up-regulated the brain BNDF protein and phosphorylation of tyrosine receptor kinase B (TrkB). Furthermore, exendin-(9–39) (a glucagon-like peptide-1 GLP-1 receptor antagonist) and K252a (a Trk tyrosine kinase inhibitor) treatment significantly abolished the cognitive protective effect of sitagliptin in the MWM test.
Sitagliptin treatment effectively protected the cognition function of the AD mice by regulating synaptic plasticity, at least partially, through activating GLP-1 and BDNF-TrkB signalings.
Finding an accurate, rapid, and automated real-time detection method for sewer is of great significance to improve the detection efficiency in sewer maintenance and ensure the system functionality. ...Deep learning technologies developed in recent years offered novel alternatives, however, most of which are computation-intensive, limiting their applications to small mobile terminals. Focusing on the characteristics and detection requirements of sewer defects, this study proposed a real-time detection method based on an integration of YOLO_v5-based object detection network, transfer learning, and channel pruning technique. Results showed that the method significantly improved the detection accuracy and speed. With the aid of model pruning, the number of parameters and operations of the original model were reduced by 81.0% and 48.8%, respectively. The model superiority was further verified by comparing with nine mainstream detection algorithms and four lightweight backbone networks. It was concluded that our model had great advantages in practical engineering applications and projects. The knowledge gained provided technical references for the deployment of portable devices and sewer inspection robots in the future.
Automatic bridge surface defect detection is of wide concern; it can save human resources and improve work efficiency. The object detection algorithm, especially the You Only Look Once (YOLO) series ...of networks, has important potential in real-time object detection because of its fast detection speed, and it provides an efficient and automatic detection method for bridge surface defect detection. Hence, this paper employs an improved YOLOv3 network for detecting bridge surface defects (cracks and exposed rebar) and compares the effects of the advanced YOLOv2, YOLOv3 and faster region-based convolutional neural network (Faster RCNN) in bridge surface defect detection, and then two approaches (transfer learning and data augmentation) are used to improve the YOLOv3. The results confirm that, by combining high- and low-resolution feature images, the YOLOv3 improves the detection effect of the YOLOv2 (using single-resolution feature images); the average precision (AP) value of the improved YOLOv3 (0.9–0.91) is 6–10% higher than that of the YOLOv2 (0.83–0.86). Then, the anti-noise abilities of the YOLOv2 and YOLOv3 are studied by introducing white Gaussian noise, and the YOLOv3 is better than the YOLOv2. Simultaneously, the YOLO series of detectors perform better in detection speed; the detection speed of the improved YOLOv3 (FPS (frames per second) = 23.8) is 103 times that of the Faster RCNN (FPS = 0.23) with comparable mAP values (improved YOLOv3 = 0.91; Faster RCNN = 0.9). It is demonstrated that, in consideration of detection precision and speed, the proposed improved YOLOv3 is a decent detector for fast and real-time bridge defect detection.
Activity-dependent insertion of the tropomyosin-related kinase B (TrkB) receptor into the plasma membrane can explain, in part, the preferential effect of brain-derived neurotrophic factor (BDNF) on ...active neurons and synapses; however, the underlying molecular mechanisms remain obscure. Here, we report a novel function for carboxypeptidase E (CPE) in controlling chemical long-term potentiation stimuli-induced TrkB surface delivery in hippocampal neurons. Total internal reflection fluorescence assays and line plot assays showed that CPE facilitates TrkB transport from dendritic shafts to the plasma membrane. The Box2 domain in the juxtamembrane region of TrkB and the C terminus of CPE are critical for the activity-dependent plasma membrane insertion of TrkB. Moreover, the transactivator of transcription TAT-CPE
, which could block the association between CPE and TrkB, significantly inhibited neuronal activity-enhanced BDNF signaling and dendritic spine morphologic plasticity in cultured hippocampal neurons. Microinfusion of TAT-CPE
into the dorsal hippocampus of male C57BL/6 mice inhibited the endogenous interaction between TrkB and CPE and diminished fear-conditioning-induced TrkB phosphorylation, which might lead to an impairment in hippocampal memory acquisition and consolidation but not retrieval. These results suggest that CPE modulates activity-induced TrkB surface insertion and hippocampal-dependent memory and sheds light on our understanding of the role of CPE in TrkB-dependent synaptic plasticity and memory modulation.
It is well known that BDNF acts preferentially on active neurons; however, the underlying molecular mechanism is not fully understood. In this study, we found that the cytoplasmic tail of CPE could interact with TrkB and facilitate the neuronal activity-dependent movement of TrkB vesicles to the plasma membrane. Blocking the association between CPE and TrkB decreased fear-conditioning-induced TrkB phosphorylation and led to hippocampal memory deficits. These findings provide novel insights into the role of CPE in TrkB intracellular trafficking as well as in mediating BDNF/TrkB function in synaptic plasticity and hippocampal memory.
This paper presents a new approach for automatical classification of structural state through deep learning. In this work, a Convolutional Neural Network (CNN) was designed to fuse both the feature ...extraction and classification blocks into an intelligent and compact learning system and detect the structural state of a steel frame; the input was a series of vibration signals, and the output was a structural state. The digital image correlation (DIC) technology was utilized to collect vibration information of an actual steel frame, and subsequently, the raw signals, without further pre-processing, were directly utilized as the CNN samples. The results show that CNN can achieve 99% classification accuracy for the research model. Besides, compared with the backpropagation neural network (BPNN), the CNN had an accuracy similar to that of the BPNN, but it only consumes 19% of the training time. The outputs of the convolution and pooling layers were visually displayed and discussed as well. It is demonstrated that: 1) the CNN can extract the structural state information from the vibration signals and classify them; 2) the detection and computational performance of the CNN for the incomplete data are better than that of the BPNN; 3) the CNN has better anti-noise ability.
PDF
