Patent
Convolutional neural network and active learning-based ceramic tile surface defect recognition method
Yao Minghai,Huang Zhancong +1 more
TLDR
In this paper, a convolutional neural network and active learning-based method was proposed for tile surface defect recognition. But the method is not suitable for the detection of tile surface defects, as good prior knowledge is not required in the aspect of defect feature extraction.Abstract:
The invention discloses a convolutional neural network and active learning-based ceramic tile surface defect recognition method. The method comprises the following steps of: (1) obtaining and preprocessing an image; (2) establishing a training set; (3) establishing and training a convolutional neural network; (4) carrying out active learning; (5) carrying out model iteration; and (6) carrying outonline detection. Compared with the prior art, the method has the advantages that (1) features of ceramic tile surface defects are automatically extracted by using the convolutional neural network, sothat good priori knowledges are not required in the aspect of defect feature extraction, and multiple defect types in one to-be-detected image can be recognized; and (2) active learning is imported in convolutional neural network training, so that labeling cost of samples is effectively decreased and the model convergence is accelerated.read more
Citations
More filters
Patent
Video frame rate upconversion method and system based on cyclic convolution neural network
TL;DR: In this paper, a video frame rate upconversion method based on a cyclic convolution neural network was proposed, which divides the original video into groups of image blocks comprising two consecutive frames of images.
Patent
Wallpaper splicing seam aligning device and wallpaper splicing seam aligning method by utilizing same
TL;DR: In this article, the authors provided a wallpaper splicing seam aligning method by utilizing the same mechanism for adjusting an included angle between the bottom surface of the wallpaper paving device and the horizontal plane under driving of the control signals, and the displacement fine adjustment mechanism was arranged on the automatic guide transportation vehicle, and was used for fine-tuning the adjustment.
Patent
Detection system and method for product surface defects
TL;DR: In this paper, a defect detection system consisting of a grasping mechanical hand, a detection mechanical hand and a control module was proposed. But the detection system was not used for the detection of product surface defects.
Patent
A method and a device for recognizing surface defects of steel based on depth learning technology
TL;DR: In this paper, a steel surface defect recognition method and a device based on depth learning technology is presented. But the method comprises the following steps of: collecting corresponding steel surface defects according to the type of steel production line; according to labeling process of the labeling platform, the defects are labeled to generate the training sample set and the verification sample set.
Patent
Universal object detection system and method for improving active learning based on self-monitoring process
TL;DR: In this paper, a universal object detection system and method for improving active learning based on a self-monitoring process is presented, which consists of a sample obtaining unit, which obtains asmall number of labeled samples and a mass of unlabeled samples, a model establishment and initialization unit, and an active learning unit which introduces the active learning course to guide the self-supervised learning process.
References
More filters
Patent
Wood defect detection method based on deep learning and system thereof
TL;DR: In this article, a wood defect detection method based on deep learning was proposed, which comprises the following steps of collecting images, segmenting the images into image blocks with the same size, selecting defect image blocks, and using the training sample set to train a deep learning algorithm in an off-line mode.
Patent
Method for detecting and classifying anomalies using artificial neural networks
TL;DR: In this article, the authors proposed a method to avoid the problem of category assignment in artificial neural networks (ANNs) based upon a mapping of the input space (like ROI and KNN algorithms) using probabilities.
Patent
Fabric defect detection method based on depth neural network
TL;DR: Wang et al. as discussed by the authors proposed a fabric defect detection method based on a depth neural network, with a convolutional neural network as a core, feature extraction is performed by a CNN, a pooling layer retains effective features and reduces the amount of calculation, and full connection layer is used for classification.
Patent
Selective search and segmentation-based diagnostic method of surface defects of fan blades
TL;DR: In this article, a selective search and segmentation-based diagnostic method of the surface defects of fan blades is proposed, and the method comprises the following steps: acquiring a candidate region from the fan blade to be shot by using selective search-and-segmentation; training a deep convolutional neural network through an ImageNet image set, and extracting networks except an output layer as an image characteristic extractor; extracting the image characteristics of a training set of the fan blades and training a support vector machine sorter; and judging the state of the patient's state to be
Related Papers (5)
Convolutional neural network-based photovoltaic glass defect classification method and device
Liu Yebin,Yan Shi,Dai Qionghai +2 more