Handwritten Number Recognition and Application Based on CNN-LSTM (with tensorflow code explanation)
abstract CNN convolution neural network is a commonly used model method in image recognition and classification.Because there is a big difference between the training effect of CNN model and the actual test, in order to improve the recognition rate of free handwritten digits, we try to construcUTF-8...
Posted by alex57 on Sat, 19 Jun 2021 03:09:16 +0930
HDL4SE: software engineers learning Verilog
11 assembly line The previous section introduced the concept of a state machine. State machine is used to describe a procedural process of transaction processing, which can be composed of sequence, branch and cycle. These concepts originally exist in the behavior level description of verilog, bUTF-8...
Posted by Cosizzle on Tue, 06 Jul 2021 04:10:50 +0930
40 cases of deep learning practical code -- CNN theory and Practice
1. General 1.1 concept Neural networks is a part of the research field of artificial intelligence. At present, the most popular neural network is deep convolutional neural networks (CNNs). Although convolutional networks also have shallow structure, they are rarely used because of accuracy and UTF-8...
Posted by berridgeab on Tue, 03 Aug 2021 07:34:53 +0930
Image classification based on convolutional neural network of PyTorch entry project
PyTorch entry project, convolutional neural network to achieve image classification task preface PyTorch is currently one of the most popular in-depth learning frameworks in academia. For students who already have the foundation of in-depth learning, there is an urgent need for a small project UTF-8...
Posted by Hatchmo on Fri, 24 Dec 2021 03:33:14 +1030
Ptorch cifar10 image classification Alex net
Ptorch cifar10 image classification Alex net 4. Define network (AlexNet) AlexNet structure: Input layer: image size is 227 × two hundred and twenty-seven × 3, where 3 indicates that the number of channel s (R, G, B) of the input image is 3. Convolution layer: filter size 11 × 11. Number of filtUTF-8...
Posted by mad_hacker on Wed, 05 Jan 2022 09:59:03 +1030
License plate recognition based on deep convolution neural network
ANPR algorithm (automatic license plate recognition system) 1. License plate detection The position of the license plate is detected in the whole video frame. 2. License plate recognition When the license plate is detected in the image, OCR (optical character recognition) algorithm is used to rUTF-8...
Posted by sirkodo on Sat, 15 Jan 2022 10:48:50 +1030
Modify StarGAN V2 using modulated revolutions
In this tutorial, we will replace the adaptive instance normalization (AdaIN) layer in the StarGAN V2 model and train it on an image with a resolution of 512x512 pixels. Why StarGAN V2? Today, there are many models that can generate high-quality images. Specifically, for the attribute exchange UTF-8...
Posted by paulb on Tue, 18 Jan 2022 00:38:41 +1030
100 cases of deep learning - convolutional neural network (CNN) attention detection | day 17
1, Preliminary work What should children do if they are always distracted in class? The attention detection program is arranged! Driver fatigue driving how to do, attention detection program arrangement! This paper will achieve the purpose of detecting attention through the recognition of humaUTF-8...
Posted by wildwobby on Thu, 20 Jan 2022 00:22:24 +1030
Target classification of MSTAR dataset based on ResNet
Target classification of MSTAR dataset based on ResNet Say in front The first two articles mainly discussed the principle of CNN https://blog.csdn.net/qq_40181592/article/details/115252377 And CNN network model https://blog.csdn.net/qq_40181592/article/details/115487195 , the main model used inUTF-8...
Posted by Hillary on Fri, 21 Jan 2022 09:51:58 +1030
Deep learning can also be played like this - 100 cases of deep learning - convolutional neural network (CNN) | day 5
1, Preliminary work In this paper, CNN will be used to identify four weather states: cloudy, rainy, sunny and sunrise. Compared with the previous article, in order to increase the generalization ability of the model, this paper adds a Dropout layer and adjusts the maximum pooling layer to the UTF-8...
Posted by AshtrayWaterloo on Wed, 09 Feb 2022 09:09:17 +1030