Xử lý ảnh trong cơ điện tử machine vision chapter 7 object recognition

TRƯỜNG ĐẠI HỌC BÁCH KHOA HÀ NỘI XỬ LÝ ẢNH TRONG CƠ ĐIỆN TỬ Machine Vision Giảng viên: TS Nguyễn Thành Hùng Đơn vị: Bộ môn Cơ điện tử, Viện Cơ khí Hà Nội, 2021 Chapter Object Recognition ❖1 Introduction ❖2 Pattern Matching ❖3 Feature-based Methods ❖4 Artificial Neural Networks Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ Object recognition: localize and to classify objects ▪ General concept: ➢ training datasets containing images with known and labelled objects; ➢ extracts different types of information (colours, edges, geometric forms) based on the chosen algorithm ➢ for any new image the same information is gathered and compared to the training dataset to find the most suitable classification Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ Applications: ➢ robots in industrial environments, ➢ face or handwriting recognition ➢ autonomous systems such as modern cars which use object recognition for pedestrian detection, emergency brake assistant and so on ➢… Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies ➢ Appearance-based method ➢ Feature-based method ➢ Interpretation Tree ➢ Pattern Matching ➢ Artificial Neural Networks Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Appearance-based method ➢ Face or handwriting recognition ➢ Reference training images ➢ This dataset is compressed to obtain a lower dimension subspace, also called eigenspace ➢ Parts of the new input images are projected on the eigenspace and then correspondence is examined Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Feature-based Method ➢ Characteristic for each object ➢ Colours, contour lines, geometric forms or edges ➢ The basic concept of feature-based object recognition strategies is following: • Every input image is searched for a specific type of feature, • This feature is then compared to a database containing models of the objects in order to verify if there are recognised objects Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Feature-based method ➢ Features and their descriptors can be either found considering the whole image (global feature) or after observing just small parts of the image (local feature) ➢ An histogram of the pixel intensity or colour are simple examples for global features ➢ It is not always reasonable to compare the whole image, as already slight changes in illumination, position (occlusion) or rotation lead to significant differences and a correct recognition is not possible anymore Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Feature-based method ➢ Descriptors of local features are more robust against these problems and therefore algorithms with local features often outperform global feature-based methods Two patches of different images are cut and compared if the error between the patches is below a certain threshold Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Interpretation Tree ➢ Interpretation tree is a depth first search algorithm for model matching ➢ Algorithms based on this approach often try to recognise n-dimensional geometric objects, therefore a database containing models with known features is necessary ➢ The feature set might consist of distance, angle and direction constraints between points on the surface of the objects Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen 10 Artificial Neural Networks ➢ The Pooling Layer Max-pooling in 3D image, which is the one that we normally deal with Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 98 Artificial Neural Networks ➢ The Pooling Layer Applying Max/Sum pooling to image that applied ReLU Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 99 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ▪ Layers used to build ConvNets ➢ Fully Connected Layer (FC) Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 100 Artificial Neural Networks ➢ Fully Connected Layer (FC) Fully connected layer Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 101 Artificial Neural Networks ➢ Fully Connected Layer (FC) ▪ Softmax function: takes a vector of arbitrary real-valued scores and squashes it to a vector of values between zero and one that sum to one Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 102 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ LeNet Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 103 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ AlexNet Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 104 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ ZFNet Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 105 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ Inception-v4 Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 106 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ VGGNet Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 107 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ VGGNet Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 108 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ ResNet Stanford University CS231n: Convolutional Neural Networks for Visual Recognition, 2020 109 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ Example: LeNet 110 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ Example: LeNet 111 Artificial Neural Networks ❖ CNN - Convolutional Neural Network ➢ Example: LeNet 112 .. .Chapter Object Recognition ❖1 Introduction ❖2 Pattern Matching ❖3 Feature-based Methods ❖4 Artificial Neural Networks Simon Achatz, State of the art of object recognition techniques,... of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Feature-based Method ➢ Characteristic for each object ➢ Colours,... verify if there are recognised objects Simon Achatz, State of the art of object recognition techniques, Technische Universitat Muchen Introduction ▪ General Object Recognition Strategies: Feature-based