One key challenging issues of facial expression recognition (FER) in video sequences is to extract discriminative spatiotemporal features from images sequences. In this paper, we propose a new method FER via hybrid deep learning model. The proposed first employs two individual convolutional neural networks (CNNs), including spatial CNN processing static and temporal CN network optical flow images, separately learn high-level on the divided segments. These CNNs are fine-tuned target datasets...

Spontaneous speech emotion recognition is a new and challenging research topic. In this paper, we propose method of spontaneous on the basis binaural representations deep convolutional neural networks (CNNs). The proposed initially employs multiple CNNs to learn segment-level such as Left-Right Mid-Side pairs from extracted image-like Mel-spectrograms. These are fine-tuned target emotional datasets pre-trained image CNN model. Then, feature pooling strategy, called block-based temporal...

Fault diagnosis for blast furnace is actually a multi-class classification problem because the may appear usually many kinds of abnormal states. Moreover, those states should be monitored and diagnosed timely what can help workers take effective measures. Support vector machine (SVM) state-of-the-art problems currently. But tasks involve imbalanced training examples in practice. Imbalanced dataset learning an important practical issue learning, especially support (SVM). such data problem. A...

Sustainable supplier selection is the essential core of sustainable supply chain management, which can directly influence manufacturer's performance and enormously enhance competitiveness in international market.However, most previous studies concerning have less focused on reliability decision-makers judgments application regret theory.To fill this gap, we presented an integrated model based theory QUALItative FLEXible multiple criteria method (QUALIFLEX) under a 2-dimensional uncertain...

The representative k-median problem generalizes the classical clustering formulations in that it partitions data points into several disjoint demographic groups and poses a lower-bound constraint on number of opened facilities from each group, such all are fairly represented by facilities. Due to its simplicity, local-search heuristic optimizes an initial solution iteratively swapping at most constant closed for same ones (denoted O(1)-swap heuristic) has been frequently used problem....

This research focuses on design and evaluation of using virtual reality (VR) serious game for general science energy education immersive learning uses the case study knowledge from calcium looping carbon capture technology. The ARCS model motivation is used in this as core promoting sustaining process. domain technology has been analyzed integrated to develop a VR game. deployment pilot 30 gifted students an elementary school collecting results by questionnaire, drawing activities, focus...

A strong state tracking particle filter (SST-PF) is put forward for unknown fault diagnosis of hybrid system. SST-PF overcomes the problem sample impoverishment nonlinear system by setting permanent transition probabilities from one mode to another. Meanwhile threshold logic normalization factor based on statistics built detect unknown-faults, which more accurate and reasonable tiny differences Simulation experiments are carried out analyze effects SST-PF, it shown that our algorithm has...

There are a wide variety of condition monitoring techniques currently used for the recognition and diagnosis machinery faults. Tool wear often results in chaotics on milling process. Little research has been carried out about occurrence detection chaotic behavior time series signal tool vibration. In paper vibration acceleration based operating stages is established analysis correlation dimension wear. Correlation calculated to recognize conditions. Finally ,some experimental from fractal...

The Wilms' tumour 1 (WT1) gene originally identified as a suppressor associated with WTs encodes zinc finger-containing transcription factor that is expressed in multiple tissues and an important regulator of cellular organ growth, proliferation, development, migration survival. However, there deficiency data regarding the expression function WT1 during oocyte maturation preimplantation embryonic development. Herein, we sought to define characteristics functions development pigs. We show...

Photovoltaic (PV) power generation is the most widely adopted renewable energy source. However, its inherent unpredictability poses considerable challenges to management of grids. To address arduous and time-consuming training process PV prediction models, which has been a major focus prior research, an improved approach for based on neighboring days proposed in this study. This specifically designed handle preprocessing datasets by leveraging results similarity analysis generation....

Face recognition has broad applications, and it is a difficult problem since face image can change with photographic conditions, such as different illumination pose changes camera angles. How to obtain some invariable features for the key issue algorithm. In this paper, novel tensor structure of proposed represent eight directions pixel value. The feature then obtained from gradient decomposition make up structure. Then singular value (SVD) principal component analysis (PCA) are used...

Locust often has similar colors and textures with the background so that it can camouflage themselves in complex surroundings. That makes difficult to identify object through common segmentation methods. PraNet performed better on colorectal polyp task, we take advantage of solve our problem. In this paper, make a new dataset Locust-mini, improve by two means, one is grouping convolution Reverse Attention(RA) module into Group Attention(GRA) which extract more meaningful information...

To strengthen the safety protection of personnel in deep water dangerous areas such as reservoirs and rivers, Relying on technical characteristics 5G network technology, high speed, low delay, bandwidth, flexibility, Build a video surveillance architecture; The cloud-edge collaboration technology based sinks cloud intelligent analysis computing power to edge side, end realizes comprehensive perception real-time data, Accelerate Form scientific effective decision-making analysis; is...

Monitoring data streams in a distributed system is the focus of much research recent years. This paper addresses generic and efficient processing top-k monitoring, which continuously reporting k largest values according to user-specified aggregation function over streams. In practice, would be arbitrary function. Unfortunately, state-of-art monitoring approaches only support sum as this paper, we present algorithm for supports not but also min, max, count, average, their compounds. These...

This paper addresses the optimal prediction for moving objects according to Pareto model. There are many factors need be taken into account when predicting objects, e.g. current position, history trajectory, weather, and so on, these often independence. Existing techniques always use a synthesized factor, which is calculated by score function contain factors, as single metric select results. Unfortunately, in applications, can't integrated one score-function, users most possible result...

Map building is very important to accomplish autonomous navigation and other complex intelligent tasks. A novel method of feature map was proposed process the uncertainty data noise sonar. This has introduced sonar arc describe detectable region for sonar, extracted line by a Hough transform based on randomized multi-resolution mobile robot. So in sensor responses can be successfully reduced. The experiment results indicate that more effective better real-time performance.