A Degradation Type Adaptive and Deep CNN-Based Image Classification Model for Degraded Images

Deep Convolutional Neural Networks(CNNs)have achieved high accuracy in image classification tasks,however,most existing models are trained on high-quality images that are not subject to image degradation.In practice,images are often affected by various types of degradation which can significantly impact the performance of CNNs.In this work,we investigate the influence of image degradation on three typical image classification CNNs and propose a Degradation Type Adaptive Image Classification Model(DTA-ICM)to improve the existing CNNs’classification accuracy on degraded images.The proposed DTA-ICM comprises two key components:a Degradation Type Predictor(DTP)and a Degradation Type Specified Image Classifier(DTS-IC)set,which is trained on existing CNNs for specified types of degradation.The DTP predicts the degradation type of a test image,and the corresponding DTS-IC is then selected to classify the image.We evaluate the performance of both the proposed DTP and the DTA-ICMon the Caltech 101 database.The experimental results demonstrate that the proposed DTP achieves an average accuracy of 99.70%.Moreover,the proposed DTA-ICM,based on AlexNet,VGG19,and ResNet152,exhibits an average accuracy improvement of 20.63%,18.22%,and 12.9%,respectively,compared with the original CNNs in classifying degraded images.It suggests that the proposed DTA-ICM can effectively improve the classification performance of existing CNNs on degraded images,which has important practical implications.

Quasi-in-situ investigation on complete lamellar fragmentation of β-solidified TiAl alloy during uniaxial isothermal compression

The coarse as-cast lamellar microstructure in TiAl alloys is difficult to be broken completely by thermomechanical processing. Some remnant lamellar colonies in the deformed microstructure seriously affect the microstructural homogeneity and deteriorate the properties. In this study, it is found that by isothermal compression at 1230 °C and 1250 °C, the lamellar colonies of Ti-43.5Al-4Nb-1Mo-0.1B (TNM) alloys can be completely broken. This is attributed to the weakened anisotropic deformation behavior of the lamellar colonies due to the isothermal holding treatment before deformation. The deformation behavior at 1230 °C was investigated by quasi-in-situ experiments. It is observed that the regions near lamellar colony boundaries first undergo dynamic recrystallization at small strain, while the lamellar colonies gradually break down with increasing strain. The adequate fragmentation of lamellar colonies mainly depends on the recrystallization of α lamellae (αL). The isothermal holding at 1230 °C leads to an increase in the content and thickness of αL, which allows it to assume more deformation and promotes its recrystallization by reaching critical strain. The interrupted γ lamellae (γL) formed by decomposition during isothermal holding facilitates the occurrence of α recrystallization within the lamellar colonies by hindering dislocation movement. In addition, recrystallized γ grains (γR) are gradually dissolved by the formation of α precipitates inside them through the γ → α phase transformation and the subsequent consumption of α precipitates by the recrystallized α grains.

A Role for Transmembrane Protein 16C/Slack Impairment in Excitatory Nociceptive Synaptic Plasticity in the Pathogenesis of Remifentanil-induced Hyperalgesia in Rats

Remifentanil is widely used to control intraoperative pain. However, its analgesic effect is limited by the generation of postoperative hyperalgesia. In this study, we investigated whether the impairment of transmembrane protein 16C(TMEM16C)/Slack is required for a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic receptor(AMPAR) activation in remifentanil-induced postoperative hyperalgesia. Remifentanil anesthesia reduced the paw withdrawal threshold from 2 h to 48 h postoperatively,with a decrease in the expression of TMEM16C and Slack in the dorsal root ganglia(DRG) and spinal cord.Knockdown of TMEM16C in the DRG reduced the expression of Slack and elevated the basal peripheral sensitivity and AMPAR expression and function. Overexpression of TMEM16C in the DRG impaired remifentanilinduced ERK1/2 phosphorylation and behavioral hyperalgesia. AMPAR-mediated current and neuronal excitability were downregulated by TMEM16C overexpression in the spinal cord. Taken together, these findings suggest that TMEM16C/Slack regulation of excitatory synaptic plasticity via GluA1-containing AMPARs is critical in the pathogenesis of remifentanil-induced postoperative hyperalgesia in rats.