Background removal and weld defect detection based on energy distribution of image

The lateral wave in ultrasonic TOFD （time of flight diffraction） image has a tail in transit time, which disturbs the detection and evaluation of shallow weld defect. Meanwhile, the lateral wave and back-wall echo that act as background add redundant data in digital image processing. In order to separate defect wave from lateral wave and prepare the way for following image processing, an algorithm of background removal method named as mean-subtraction is developed. Based on this, an improved method by statistic of the energy distribution in the image is proposed. The results show that by choosing proper threshold value according to the axial energy distribution of the image, the background can be removed automatically and the defect section becomes predominant. Meanwhile, diffractive wave of shallow weld defect can be separated from lateral wave effectively.

Patch-based vehicle logo detection with patch intensity and weight matrix

A patch-based method for detecting vehicle logos using prior knowledge is proposed.By representing the coarse region of the logo with the weight matrix of patch intensity and position,the proposed method is robust to bad and complex environmental conditions.The bounding-box of the logo is extracted by a thershloding approach.Experimental results show that 93.58% location accuracy is achieved with 1100 images under various environmental conditions,indicating that the proposed method is effective and suitable for the location of vehicle logo in practical applications.

CRISPR-detector:fast and accurate detection,visualization,and annotation of genome-wide mutations induced by genome editing events

The leading-edge CRISPR/CRISPR-associated technology is revolutionizing biotechnologies through genome editing.To track on/off-target events with emerging new editing techniques,improved bioinformatic tools are indispensable.Existing tools suffer from limitations in speed and scalability,especially with whole-genome sequencing(WGS)data analysis.To address these limitations,we have developed a comprehensive tool called CRISPR-detector,a web-based and locally deployable pipeline for genome editing sequence analysis.The core analysis module of CRISPR-detector is based on the Sentieon TNscope pipeline,with additional novel annotation and visualization modules designed to fit CRISPR applications.Co-analysis of the treated and control samples is performed to remove existing background variants prior to genome editing.CRISPR-detector offers optimized scalability,enabling WGS data analysis beyond Browser Extensible Data file-defined regions,with improved accuracy due to haplotype-based variant calling to handle sequencing errors.In addition,the tool also provides integrated structural variation calling and includes functional and clinical annotations of editing-induced mutations appreciated by users.These advantages facilitate rapid and efficient detection of mutations induced by genome editing events,especially for datasets generated from WGS.The web-based version of CRISPR-detector is available at https://db.cngb.org/crispr-detector,and the locally deployable version is available at https://github.com/hlcas/CRISPR-detector.

A new approach to learning and recognizing leaf diseases from individual lesions using convolutional neural networks

Leaf disease recognition using image processing and deep learning techniques is currently a vibrant research area.Most studies have focused on recognizing diseases from images of whole leaves.This approach limits the resulting models’ability to estimate leaf disease severity or identify multiple anomalies occurring on the same leaf.Recent studies have demonstrated that classifying leaf diseases based on individual lesions greatly enhances disease recognition accuracy.In those studies,however,the lesions were laboriously cropped by hand.This study proposes a semi-automatic algorithm that facilitates the fast and efficient preparation of datasets of individual lesions and leaf image pixel maps to overcome this problem.These datasets were then used to train and test lesion classifier and semantic segmentation Convolutional Neural Network(CNN)models,respectively.We report that GoogLeNet’s disease recognition accuracy improved by more than 15%when diseases were recognized from lesion images compared to when disease recognition was done using images of whole leaves.A CNN model which performs semantic segmentation of both the leaf and lesions in one pass is also proposed in this paper.The proposed KijaniNet model achieved state-of-the-art segmentation performance in terms of mean Intersection over Union(mIoU)score of 0.8448 and 0.6257 for the leaf and lesion pixel classes,respectively.In terms of mean boundary F1 score,the KijaniNet model attained 0.8241 and 0.7855 for the two pixel classes,respectively.Lastly,a fully automatic algorithm for leaf disease recognition from individual lesions is proposed.The algorithm employs the semantic segmentation network cascaded to a GoogLeNet classifier for lesion-wise disease recognition.The proposed fully automatic algorithm outperforms competing methods in terms of its superior segmentation and classification performance despite being trained on a small dataset.