Search

With the development of computer technology, fire image processing technology combining computer vision, machine learning, deep learning and other technologies has been widely studied and applied. Aiming at the complex preprocessing process and high false positive rate of traditional image processing methods, this paper proposes a method based on deep convolutional neural network model for fire detection, which reduces complex preprocessing links and integrates the whole fire identification process into one single depth neural network for easy training and optimization. In view of the problem of fire detection caused by similar fire scenes in the identification process, this paper uses the motion characteristics of fire to innovatively propose the combination of fire frame position changes before and after the fire video to eliminate the interference of lights and other similar fire scenes. After comparing many open learning open source frameworks, this paper chooses Caffe framework for training and testing. The experimental results show that the method realizes the recognition and localization of fire images. This method is suitable for different fire scenarios and has good generalization ability and anti-interference ability.

This paper propose a finger vein recognition algorithm based on the CapsNets(Capsule Network for short) to solve the problem of the information loss of the finger vein in the Convolution Neural Network(CNN). The CapsNets is transferred from the bottom to the high level in the form of capsule in the whole learning process, so that the multidimensional characteristics of the finger vein are encapsulated in the form of vector, and the features will be preserved in the network, but not in the network after the loss is recovered. In this paper, 60 000 images are used as training set, and 10 000 images are used as test set. The experimental results show that the network structure features of CapsNets are more obvious than that of CNN, the accuracy of VGG is increased by 13.6%, and the value of loss converges to 0.01.

At present, accurate face recognition has been widely used in real scenes, such as monitoring, access control, etc., and recognition only for face attributes is also widely used in entertainment, security and social media. Therefore, the paper is based on deep convolutional neural network MTCNN for face detection and Resnet50 network for face attribute recognition. At the same time, it uses Python language to design and develop a system that implements face attribute recognition application interface. Through data experiments, the system uses a face attribute recognition model based on deep convolutional neural networks to achieve 97.32%, 71.64% and 92.13% accuracy in identifying gender, age and ethnic attributes, respectively. At the same time, the face attribute recognition application system can run stably and has certain application value.

This paper proposes to apply Transformer model in the field of Chinese text automatic proofreading. Transformer model is different from traditional Seq2Seq model based on probability, statistics, rules or BiLSTM. This deep learning model improves the overall structure of Seq2Seq model to achieve automatic proofreading of Chinese text. By comparing different models with public data sets and using accuracy, recall rate and F1 value as evaluation indexes, the experimental results show that Transformer model has greatly improved proofreading performance compared with other models.

Crop yield forecasting plays a decisive role in the development of national economy planned by the government. It is of great significance to rationally plan planting strategies and reduce the waste of water and fertilizer. There are many factors affecting crop yield, and accurate prediction of crop growth means a lot. Climate is an important factor affecting crop yield. Based on climatic factors, this paper proposes a time series prediction method based on the long short-term memory(LSTM) for crop yield prediction. The method combines historical yields with climatic factors to forecast the crop yield of the next period by years. Experimental results show that compared with LSTM and support vector machine methods, the proposed method achieves higher accuracy in crop yield prediction.

Identity authentication technology has developed greatly, and there have been various fraudulent means of forging legitimate user information. Aiming at this problem, this paper proposes a deep learning face detection algorithm to analyze the difference between real face and fraud face, decentralize the real face and photo, zca whiten to noise, random rotation and other processing. At the same time, using the convolutional neural network to extract the facial features of the photos, the extracted features are sent to the neural network for training and classification. And the algorithm is verified on the public database NUAA. The experimental results show that the party reduces the calculation complexity and increases the recognition accuracy.

With the advent of the “artificial intelligence” era, the term “deep learning” has gradually entered the public′s field of vision. Some image processing methods based on deep learning neural networks have also emerged. Image stylization as an important branch has also gained widespread attention. At present, researchers have proposed a lot of image stylization algorithms based on deep learning, and they can accomplish stylized tasks well. This paper comprehensively summarizes the progress of deep learning in the field of image stylization, compares the advantages and disadvantages of different algorithms, and finally discusses the limitations of current image stylization research based on deep learning and future research directions.

Modulation recognition of radio signals plays a vital role in radio monitoring and spectrum management. As the deep learning network in artificial neural network has the powerful ability of representation learning which can automatically extract various complex features from the original data, exploring the modulation identification of radio signals based on deep learning is one of the main development trends in the field of radio monitoring. This paper introduces some application results and existing problems of deep learning in radio signal modulation recognition. Combined with the actual needs of the work, this review puts forward some ideas for deep learning in the modulation recognition of radio signals, such as further improving the recognition range and the recognition accuracy, especially at low SNR; seeking some new deep learning hybrid architecture for radio signal modulation recognition.

With the rise of deep learning in recent years, it has made major breakthroughs in machine learning fields such as target detection, image classification, speech recognition, and natural language processing. Among them, convolutional neural networks are widely used in deep learning. Since the emergence of VGGNet, neural networks have gradually developed deeper and the network has become deeper. This not only increases the demand for operating memory and running memory, but also greatly increases the amount of computation. The requirements for hardware platform resources are getting higher and higher. Therefore, it is particularly difficult to apply deep learning to embedded platforms. In this paper, by pruning the model, the trained network model is compressed, the unimportant parameters are eliminated, the model is reduced, and the computational complexity of the network is reduced. The deep learning will be applied to the embedded platform.

This paper analyses the hot point about the “Belt and Road” initiative of American mainstream news media and studies the sentiment of related public opinion. Web crawler is used to automatically collect relevant news and filter high-frequency words to get media attention hotspots. An integrated model of automatic summary-convolutional neural network(CNN) is proposed for document-level sentiment analysis. The model firstly extracts the abstraction to remove the interference of non-important data in the original document, then the convolutional neural network is used to analyze the sentence-level sentiment, obtain the document-level emotional score based on the semantic pointing method, and the emotional fluctuation abnormal articles are analyzed twice. Contrastive experiments on real data shows that the automatic summary-CNN integrated document-level sentiment analysis model is superior to the single CNN method in sentiment analysis.

The recognition of handwritten digits is an important part of the artificial intelligence recognition system. Due to the difference in individual handwritten numbers, the existing recognition system has a lower accuracy rate. This paper is based on the TensorFlow deep learning framework to complete the recognition and application of handwritten numbers. Firstly, the Softmax and Convolutional Neural Network(CNN) model structure is established and analyzed. Secondly, deep learning is performed on 60 000 samples of the handwritten data set MNIST, and then 10 000 samples are tested and compared. Finally, the optimal model is transplanted to the Android platform for application. Compared with the traditional Softmax model, the recognition rate based on TensorFlow deep learning CNN model is as high as 99.17%, an increase of 7.6%, which provides certain scientific research value for the development of artificial intelligence recognition system.

This paper develops a strawberry robot-arm based on computer vision, and uses deep learning to localize targets for strawberry robot-arm. In order to speed up localization, the algorithm segments the image from strawberry robot-arm based on color prior knowledge to get strawberry proposals, then inputs proposals to convolutional neural network(CNN) pretrained and gets the probability for every proposal, and finally chooses the target with the highest probability as the localization result. Experiment results show that this algorithm can localize target in real time and with high precision.

In order to improve the problem of low accuracy in human behavior recognition task, a neural network based on batch normalization convolution neural network(CNN) and long short-term memory(LSTM) neural network is proposed. The CNN part introduces the idea of batch normalization, and the training data of the input network are normalized in mini-batch. After full connection, they are sent to long short-term memory neural network. The algorithm adopts the space-time dual stream network model structure. The RGB image of video data is taken as spatial stream network input, and the optical flow field image is taken as time flow network input. Then the recognition results obtained by the time-space dual-stream network are combined in a certain proportion to obtain the final behavior recognition result. The experimental results show that the space-time dual stream neural network algorithm designed in this paper has a high recognition accuracy in human behavior recognition tasks.

This paper designs a real-time recognition hardware system framework based on deep learning. The system framework uses Keras to complete the training of the convolutional neural network model and extracts the parameters of the network. Using the FPGA+ARM software and hardware coordination method of the ZYNQ device, ARM was used to complete the acquisition, preprocessing and display of real-time image data. Through the FPGA，the hardening of the convolutional neural network is performed and the image is recognized, and the recognition result is sent to the upper computer for real-time display. The system framework uses MNIST and Fashion MNIST data sets as network model hardening test samples. The experimental results show that the system framework can display and identify image data in real time and accurately under the general scene. And it has the characteristics of high portability, fast processing speed and low power consumption.

This paper proposed a scheme design of intelligent road card recognition system based on machine vision and deep learning. ARM9 is used as the part of front-end collection system,the image processing algorithm is used to complete the area extraction and segmentation of the road card text from the front-end acquisition image, the deep learning with a convolutional neural network is used to recognize the text in the cloud server, and the speech of the text information is presented to the user. Using the front-end hardware to get the road card images on the high way, the function of our system under CAFFE framework on the server is tested and the results show that the system can broadcast the road card information effectively to the user in real time.

This paper proposes a convolution neural network（CNN） for image classification, which uses overlap pooling and dropout technology to solve overfitting problem. Compared with traditional CNN，the proposal obtains better results on the CIFAR-10 dataset，where the accuracy on testing data set is about 9 percent higher than that on the training data set.

In this paper, a deep convolution neural network system is designed and implemented by FPGA hardware platform for the problem that the convolution neural network(CNN) in deep learning is slow and time consuming under the CPU platform. The system uses the rectified linear unit(ReLU) as the characteristic output activation function and uses the Softmax function as the output classifier. Assembly line and the parallelism are used for the feature operation of each layer, so that 295 convolution operations in the entire CNN can be completed in one system clock cycle. The system finally uses the MNIST data set as the experimental, and experimental results show that the training time of FPGA work on 50 MHz is 8.7 times higher than that of general-purpose CPU, and the accuracy rate of system identification after 2 000 iterations is 92.42%.

This paper proposes an intelligent automated vending system based on machine vision and deep learning. The embedded ARM9 and various sensors are used to form the front-end acquisition system. The trained convolutional neural network is used to detect and identify the goods dynamically and statically on the server. Then the identifying information will be fed back to the databases, and the databases will integrate all the information. Finally it will determine the order information of customers. This system uses front-end hardware to acquire images on the automated vending counter and test them on the server's Caffe framework, and the result shows that the accuracy of this system can reach 99% in real time.