ACF Based Region Proposal Extraction for YOLOv3 Network Towards High-Performance Cyclist Detection in High Resolution Images

Automatic identification of medically important mosquitoes using embedded learning approach-based image-retrieval system

Mosquito-borne diseases such as dengue fever and malaria are the top 10 leading causes of death in low-income countries. Control measure for the mosquito population plays an essential role in the fight against the disease. Currently, several intervention strategies; chemical-, biological-, mechanical- and environmental methods remain under development and need further improvement in their effectiveness. Although, a conventional entomological surveillance, required a microscope and taxonomic key for identification by professionals, is a key strategy to evaluate the population growth of these mosquitoes, these techniques are tedious, time-consuming, labor-intensive, and reliant on skillful and well-trained personnel. Here, we proposed an automatic screening, namely the deep metric learning approach and its inference under the image-retrieval process with Euclidean distance-based similarity. We aimed to develop the optimized model to find suitable miners and suggested the robustness of the proposed model by evaluating it with unseen data under a 20-returned image system. During the model development, well-trained ResNet34 are outstanding and no performance difference when comparing five data miners that showed up to 98% in its precision even after testing the model with both image sources: stereomicroscope and mobile phone cameras. The robustness of the proposed-trained model was tested with secondary unseen data which showed different environmental factors such as lighting, image scales, background colors and zoom levels. Nevertheless, our proposed neural network still has great performance with greater than 95% for sensitivity and precision, respectively. Also, the area under the ROC curve given the learning system seems to be practical and empirical with its value greater than 0.960. The results of the study may be used by public health authorities to locate mosquito vectors nearby. If used in the field, our research tool in particular is believed to accurately represent a real-world scenario.

Deep learning approaches for challenging species and gender identification of mosquito vectors

Microscopic observation of mosquito species, which is the basis of morphological identification, is a time-consuming and challenging process, particularly owing to the different skills and experience of public health personnel. We present deep learning models based on the well-known you-only-look-once (YOLO) algorithm. This model can be used to simultaneously classify and localize the images to identify the species of the gender of field-caught mosquitoes. The results indicated that the concatenated two YOLO v3 model exhibited the optimal performance in identifying the mosquitoes, as the mosquitoes were relatively small objects compared with the large proportional environment image. The robustness testing of the proposed model yielded a mean average precision and sensitivity of 99% and 92.4%, respectively. The model exhibited high performance in terms of the specificity and accuracy, with an extremely low rate of misclassification. The area under the receiver operating characteristic curve (AUC) was 0.958 ± 0.011, which further demonstrated the model accuracy. Thirteen classes were detected with an accuracy of 100% based on a confusion matrix. Nevertheless, the relatively low detection rates for the two species were likely a result of the limited number of wild-caught biological samples available. The proposed model can help establish the population densities of mosquito vectors in remote areas to predict disease outbreaks in advance.

A PCB Electronic Components Detection Network Design Based on Effective Receptive Field Size and Anchor Size Matching

Vision-based recognizing and positioning of electronic components on the PCB (printed circuit board) can improve the quality inspection efficiency of electronic products in the manufacturing process. With the improvement of the design and the production process, the electronic components on the PCB show the characteristics of small sizes and similar appearances, which brings challenges to visual object detection. This paper designs a real-time electronic component detection network through effective receptive field size and anchor size matching in YOLOv3. We make contributions in the following three aspects: (1) realizing the calculation and visualization of the effective receptive field size of the different depth layers of the CNN (convolutional neural network) based on gradient backpropagation; (2) proposing a modular YOLOv3 composition strategy that can be added and removed; and (3) designing a lightweight and efficient detection network by effective receptive field size and anchor size matching algorithm. Compared with the Faster-RCNN (regions with convolutional neural network) features, SSD (single-shot multibox detectors), and original YOLOv3, our method not only has the highest detection mAP (mean average precision) on the PCB electronic component dataset, which is 95.03%, the smallest parameter size of the memory, about 1/3 of the original YOLOv3 parameter amount, but also the second-best performance on FLOPs (floating point operations).

Multi-Scale and Occlusion Aware Network for Vehicle Detection and Segmentation on UAV Aerial Images

With the advantage of high maneuverability, Unmanned Aerial Vehicles (UAVs) have been widely deployed in vehicle monitoring and controlling. However, processing the images captured by UAV for the extracting vehicle information is hindered by some challenges including arbitrary orientations, huge scale variations and partial occlusion. In seeking to address these challenges, we propose a novel Multi-Scale and Occlusion Aware Network (MSOA-Net) for UAV based vehicle segmentation, which consists of two parts including a Multi-Scale Feature Adaptive Fusion Network (MSFAF-Net) and a Regional Attention based Triple Head Network (RATH-Net). In MSFAF-Net, a self-adaptive feature fusion module is proposed, which can adaptively aggregate hierarchical feature maps from multiple levels to help Feature Pyramid Network (FPN) deal with the scale change of vehicles. The RATH-Net with a self-attention mechanism is proposed to guide the location-sensitive sub-networks to enhance the vehicle of interest and suppress background noise caused by occlusions. In this study, we release a large comprehensive UAV based vehicle segmentation dataset (UVSD), which is the first public dataset for UAV based vehicle detection and segmentation. Experiments are conducted on the challenging UVSD dataset. Experimental results show that the proposed method is efficient in detecting and segmenting vehicles, and outperforms the compared state-of-the-art works.

Cyclist detection and tracking based on multi-layer laser scanner

The technology of Artificial Intelligence (AI) brings tremendous possibilities for autonomous vehicle applications. One of the essential tasks of autonomous vehicle is environment perception using machine learning algorithms. Since the cyclists are the vulnerable road users, cyclist detection and tracking are important perception sub-tasks for autonomous vehicles to avoid vehicle-cyclist collision. In this paper, a robust method for cyclist detection and tracking is presented based on multi-layer laser scanner, i.e., IBEO LUX 4L, which obtains four-layer point cloud from local environment. First, the laser points are partitioned into individual clusters using Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method based on subarea. Then, 37-dimensional feature set is optimized by Relief algorithm and Principal Component Analysis (PCA) to produce two new feature sets. Support Vector Machine (SVM) and Decision Tree (DT) classifiers are further combined with three feature sets, respectively. Moreover, Multiple Hypothesis Tracking (MHT) algorithm and Kalman filter based on Current Statistical (CS) model are applied to track moving cyclists and estimate the motion state. The performance of the proposed cyclist detection and tracking method is validated in real road environment.

Visual Locating of Reactor in an Industrial Environment Using the Composite Method

To achieve an automatic unloading of a reactor during the sherardizing process, it is necessary to calculate the pose and position of the reactors in an industrial environment with various amounts of luminance and floating dust. In this study, the defects of classic image processing methods and deep learning methods used for locating the reactors are first analyzed. Next, an improved You Only Look Once(YOLO) model is employed to find the region of interest of the handling hole and a handling hole corner detection method based on the image morphology and a Hough transform is presented. Finally, the position and pose of the reactors will be obtained by establishing a 3D handling hole model according to the principle of a binocular stereo system. To test the performance of the proposed method, a set of experimental systems was set up and experiments were conducted. The results indicate that the proposed location method is effective and the precision of the position recognition can be controlled to within 4.64 mm and 1.68° when the cameras are approximately 5 m away from the reactor, meeting the requirements.