Improved Faster R-CNN Research Articles

Early Bruise Detection in Apple Based on an Improved Faster RCNN Model

Bruising is a common occurrence in apples that can lead to gradual fruit decay and substantial economic losses. Due to the lack of visible external features, the detection of early-stage bruising (occurring within 0.5 h) is difficult. Moreover, the identification of stems and calyxes is also important. Here, we studied the use of the short-wave infrared (SWIR) camera and the Faster RCNN model to enable the identification of bruises on apples. To evaluate the effectiveness of early bruise detection by SWIR bands compared to the visible/near-infrared (Vis/NIR) bands, a hybrid dataset with images from two cameras with different bands was used for validation. To improve the accuracy of the model in detecting apple bruises, calyxes, and stems, several improvements are implemented. Firstly, the Feature Pyramid Network (FPN) structure was integrated into the ResNet50 feature extraction network. Additionally, the Normalization-based Attention Module (NAM) was incorporated into the residual network, serving to bolster the attention of model towards detection targets while effectively mitigating the impact of irrelevant features. To reduce false positives and negatives, the Intersection over Union (IoU) metric was replaced with the Complete-IoU (CIoU). Comparison of the detection performance of the Faster RCNN model, YOLOv4P model, YOLOv5s model, and the improved Faster RCNN model, showed that the improved model had the best evaluation indicators. It achieved a mean Average Precision (mAP) of 97.4% and F1 score of 0.87. The results of research indicate that it is possible to accurately and effectively identify early bruises, calyxes, and stems on apples using SWIR cameras and deep learning models. This provides new ideas for real-time online sorting of apples for the presence of bruises.

Identification of Lunar Craters in the Chang’e-5 Landing Region Based on Kaguya TC Morning Map

Impact craters are extensively researched geological features that contribute to various aspects of lunar science, such as evaluating the model age, regolith thickness, etc. The method for identifying impact craters has gradually transitioned from manual counting to automated identification. Automatic crater detection based on the digital elevation model (DEM) is commonly used to detect larger craters. However, using only DEM has limitations in discerning smaller craters (diameter < ~1 km). This study utilizes an improved Faster R-CNN algorithm and the Kaguya Terrain Camera (TC) morning map to detect small impact craters in the Chang’e-5 (CE-5) landing site. It uses model fusion to improve the precision of small crater identification. The results show a recall rate of 96.33% and a precision value of 90.19% for craters with diameters exceeding 200 m. The model found a total of 187,101 impact craters in the CE-5 region. The spatial distribution density of impact craters with diameters ranging from 100 m to 200 m is approximately 2.5706/km2. For craters with diameters ranging from 200 m to 1 km, the average spatial distribution density is about 0.9016/km2. By the unbiased impact crater density of chronological analysis, the model age of the Im2 and Em4 geological units in the CE-5 region is 3.78 Ga and 2.07 Ga, respectively.

Detection of Water Meter Digits Based on Improved Faster R-CNN

Detection of Water Meter Digits Based on Improved Faster R-CNN

Martian Dust Devil Detection Based on Improved Faster R-CNN

Martian Dust Devil Detection Based on Improved Faster R-CNN

Infrared target detection algorithm based on improved Faster R-CNN

Infrared target detection algorithm based on improved Faster R-CNN

Vehicle Target Detection Algorithm Based on Improved Faster R-CNN for Remote Sensing Images

Vehicle Target Detection Algorithm Based on Improved Faster R-CNN for Remote Sensing Images

Research on Pedestrian Detection Based on Multimodal Infor-mation Fusion

Aiming at the matter that pedestrian detection in the autonomous driving system is vulnerable to the influence of the external environment and the detector supported single sensor modal detector has poor performance beneath the condition of enormous amendment of unrestricted light-weight, this paper proposes a fusion of light and thermal infrared dual mode pedestrian detection methodology. Firstly, 1 × 1 convolution and expanded convolution square measure are introduced within the residual network, and also the ROI Align methodology is employed to exchange the ROI Pooling method-ology to map the candidate box to the feature layer to optimize the Faster R-CNN. Secondly, the loss performance of the generalized intersection over union (GIoU) is employed because of the loss performance of the prediction box positioning regression; finally, supported by the improved Faster R-CNN, four forms of multimodal neural network structures square measure designed to fuse visible and thermal infrared pictures. According to experimental findings, the proposed technique outperforms current mainstream detection algorithms on the KAIST dataset. As compared to the conventional ACF + T + THOG pedestrian detector, the AP is 8.38 percentage points greater. Compared to the visible light pedestrian detector, the miss rate is 5.34 percentage points lower.

Small lunar crater identification and age estimation in Chang'e-5 landing area based on improved Faster R-CNN

The Chang'e-5 (CE-5) mission marks China's first lunar sample return endeavor, with its landing site (43.06°N, 51.92°W) situated in the Mons Rümker region of the northern Oceanus Procellarum on the Moon. This region hosts some of the youngest mare basalts of the Moon and contains a relatively youthful geologic unit characterized by crater's equilibrium diameters slightly over 100 m. By refining the Faster Region-based Convolutional Neural Network (Faster R-CNN) algorithm and leveraging high-resolution imagery to create training samples, accurate identification of lunar craters can be achieved. In this study, we enhance the algorithm in aspects such as anchor boxes and Region of Interest alignment. Additionally, we have utilized high-resolution images for training, and identify and statistics craters within the CE-5 landing area. Ultimately, our model attains a validation set Recall of 90%, Precision of 69%, and an Average Precision score of 0.83. Notably, in certain scales, such as for the crater larger than 400 m, recognition results reach Precision of 88% and Recall of 89%. The findings of this study are mapped into a crater catalog. Furthermore, we predict crater density and integrate it with geochronological functions to estimate the absolute model age of nine major geologic units within the CE-5 landing area. The results are generally in agreement with those of other studies who have used manual methods for crater counting, and verify the correctness of our automatic crater identification results.

Image Object Detection Method Based on Improved Faster R-CNN

Aiming at the problems about the incompleteness and low accuracy in classifying image feature extraction in existing image object detection methods. First, an improved image object detection method based on faster Region-Convolutional Neural Network (RCNN) is proposed, and Region of Interest (ROI) align is used instead of RoI pooling to reduce the error in the pooling process. Second, it adds the layer of convolution before adding network full-connection layer, thus reducing its parameters, enhance the performance of the classifier and avoid over-fitting. Finally, the combination of softmax and central loss functions is for training network-based model to increase the differences between categories and reduce the changes within categories, and thus the model with network could have more targeted and diversified features. Research results found that in total, method average accuracy is 91.04%, indicating higher than the accuracy of other methods and has good image target detection ability.

An Improved Faster R-CNN Method for Landslide Detection in Remote Sensing Images

An Improved Faster R-CNN Method for Landslide Detection in Remote Sensing Images

Improved Faster R-CNN Traffic Sign Detection Algorithm Based on Transformer

In view of the fact that traffic sign detection is becoming more and more important in assisted driving, but there are situations where the target is small, occluded and the background is complex, this paper proposes a Transformer-based improved Faster R-CNN algorithm for traffic sign detection. The algorithm in this paper uses the Transformer network based on the shift window as the backbone. The multi-level feature map is fused across layers through the cascade fusion module to obtain the fusion feature map. Moreover, the model’s feature extraction ability is improved. The detection head module eliminates rounding through RoI Align quantization error. Based on experiments in the TT100K dataset, it can be known that the mAP and detection speed of the algorithm in this paper has been improved. The effectiveness of the improved model in this paper is proven and it is applicable to real scenarios.

Research on fabric yarn detection based on improved fast R-CNN algorithm

Abstract For the current commercially available fabric yarn detection, which is based on traditional machine vision methods and relies heavily on manually designed features, an improved Faster R-CNN algorithm is proposed in this paper. In this paper, based on the Faster R-CNN algorithm, the deformable convolutional Resnet-50 network is fused to improve the learning ability of woven yarn features. By designing a multi-scale model for the detection of fine features in fabric yarns, a cascade network is introduced to improve the detection accuracy and localization accuracy of woven yarns, and an optimized loss function is constructed to reduce the effect of sample imbalance. Through the experiments of the improved fast R-CNN algorithm for fabric yarn detection, we can find that the confidence level of SSD algorithm detection is 58%, and the confidence level of the original Faster R-CNN algorithm detection is 78%, while the improved Faster R-CNN can not only accurately frame the finesse problem, but also the confidence level is as high as 97%. So this paper, based on the improved Faster R-CNN algorithm, can pinpoint the problem of fabric yarn detection, improve the learning ability of detection, and can meet the current demand for fabric yarn detection in the market.

Multi-Target Rumination Behavior Analysis Method of Cows Based on Target Detection and Optical Flow Algorithm

Rumination behavior is closely associated with factors such as cow productivity, reproductive performance, and disease incidence. For multi-object scenarios of dairy cattle, ruminant mouth area images accounted for little characteristic information, which was first put forward using an improved Faster R-CNN target detection algorithm to improve the detection performance model for the ruminant area of dairy cattle. The primary objective is to enhance the model’s performance in accurately detecting cow rumination regions. To achieve this, the dataset used in this study is annotated with both the cow head region and the mouth region. The ResNet-50-FPN network is employed to extract the cow mouth features, and the CBAM attention mechanism is incorporated to further improve the algorithm’s detection accuracy. Subsequently, the object detection results are combined with optical flow information to eliminate false detections. Finally, an interpolation approach is adopted to design a frame complementary algorithm that corrects the detection frame of the cow mouth region. This interpolation algorithm is employed to rectify the detection frame of the cow’s mouth region, addressing the issue of missed detections and enhancing the accuracy of ruminant mouth region detection. To overcome the challenges associated with the inaccurate extraction of small-scale optical flow information and interference between different optical flow information in multi-objective scenes, an enhanced GMFlowNet-based method for multi-objective cow ruminant optical flow analysis is proposed. To mitigate interference from other head movements, the MeanShift clustering method is utilized to compute the velocity magnitude values of each pixel in the vertical direction within the intercepted ruminant mouth region. Furthermore, the mean square difference is calculated, incorporating the concept of range interquartile, to eliminate outliers in the optical flow curve. Finally, a final filter is applied to fit the optical flow curve of the multi-object cow mouth movement, and it is able to identify rumination behavior and calculate chewing times. The efficacy, robustness, and accuracy of the proposed method are evaluated through experiments, with nine videos capturing multi-object cow chewing behavior in different settings. The experimental findings demonstrate that the enhanced Faster R-CNN algorithm achieved an 84.70% accuracy in detecting the ruminant mouth region, representing an improvement of 11.80 percentage points over the results obtained using the Faster R-CNN detection approach. Additionally, the enhanced GMFlowNet algorithm accurately identifies the ruminant behavior of all multi-objective cows, with a 97.30% accuracy in calculating the number of ruminant chewing instances, surpassing the accuracy of the FlowNet2.0 algorithm by 3.97 percentage points. This study provides technical support for intelligent monitoring and analysis of rumination behavior of dairy cows in group breeding.

Aircraft Detection from Low SCNR SAR Imagery Using Coherent Scattering Enhancement and Fused Attention Pyramid

Recently, methods based on deep learning have been applied to target detection using synthetic aperture radar (SAR) images. However, due to the SAR imaging mechanism and low signal-clutter-noise-ratio (SCNR), it is still a challenging task to perform aircraft detection using SAR imagery. To address this issue, a novel aircraft detection method is proposed for low SCNR SAR images that is based on coherent scattering enhancement and a fusion attention mechanism. Considering the scattering characteristics discrepancy between human-made targets and natural background, a coherent scattering enhancement technique is introduced to heighten the aircraft scatter information and suppress the clutter and speckle noise. This is beneficial for the later ability of the deep neural network to extract accurate and discriminative semantic information about the aircraft. Further, an improved Faster R-CNN is developed with a novel pyramid network constructed by fusing local and contextual attention. The local attention adaptively highlights the significant objects by enhancing their distinguishable features, and the contextual attention facilitates the network to extract distinct contextual information of the image. Fusing the local and contextual attention can guarantee that the aircraft is detected as completely as possible. Extensive experiments are performed on TerraSAR-X SAR datasets for benchmark comparison. The experimental results demonstrate that the proposed aircraft detection approach could achieve up to 91.7% of average precision in low SCNR, showing effectiveness and superiority over a number of benchmarks.

Square ancient sites detection in typical regions of the Mongolian plateau using improved faster R-CNN from Google Earth high-resolution images

ABSTRACT Many square ancient sites exist in the Mongolian plateau region, and locating them is of great importance for archaeological research. In recent years, object detection in Google Earth (GE) high-resolution remote sensing (RS) images based on deep learning has driven archaeologists’ research of ancient site detection. Deep learning techniques significantly improve the accuracy of site detection in RS images when compared with traditional detection methods while reducing design and computation time. Based on the characteristics of the research objectives, this paper proposes an improved Faster R-CNN algorithm for the detection of square ancient sites in GE high-resolution RS images, which employs Swin-Transformer, VGG16, Atrous Spatial Pyramid Pooling (ASPP), Squeeze-and-Excitation Networks (SENet) to form an improved backbone network, and three methods to optimize the Region Proposal Network (RPN). In addition, a dataset of known square ancient sites containing four typical regions of the Mongolian plateau is built using GE high-resolution RS images for training and evaluating the algorithm model. The experimental results show that the Precision, Recall, F1, IoU, and Average Precision (AP) of the proposed improved Faster R-CNN are 91.38%, 91.16%, 91.27%, 83.49%, and 93.57%, respectively, which are 34.89%, 2.70%, 22.29%, 31.46%, and 7.32% higher than those of the original Faster R-CNN. The algorithm also has high evaluation metrics for object detection in four typical regions: hills, Gobi, cropland, and grassland. Finally, the proposed algorithm is applied to detect square ancient sites in six selected typical regions, and numerous new ones are discovered. Overall, the proposed algorithm provides an effective method for archaeological investigations in the study region.

Tile Surface Defect Detection Based on Improved Faster R-CNN

With the increasing population, there is a high demand for housing, commercial buildings, and lifestyle facilities. The development of modern manufacturing technology has led to higher quality control requirements for tiles. This article presents a ceramic tile defect detection method based on Faster R-CNN, primarily focused on detecting surface defects or cracks on tiles. The method replaces VGG16 with the ResNet-101 network as a new backbone and utilizes depth-wise separable convolution to address efficiency issues resulting from the backbone replacement. Finally, Soft-NMS is employed to optimize regression boxes and prevent missed detections. Experimental results demonstrate that the improved algorithm achieves a mAP of 70.4%, a 13.2% enhancement over the original algorithm, highlighting the effectiveness and feasibility of the proposed approach.

Improved Faster R-CNN Network for Liquid Bag Foreign Body Detection

The production quality of medical fluid bags is closely related to patient health. In this paper, we used medical fluid bags to detect whether they contained foreign bodies. A visual acquisition system for the fluid bag was built. Vignetting correction was performed on the acquired images, and a foreign body recognition detection method based on an improved Faster R-CNN model was proposed. The feature extraction network of Faster R-CNN was discussed and studied regarding the characteristics of small foreign objects in liquid bags, and the ResNet152 network replaced the VGG16 network; furthermore, the feature fusion and attention mechanism were added to the feature extraction, and CIoU replaced the IoU loss function; the anchor box parameters were optimized and improved using the K-means clustering algorithm, and ROI Align replaced the ROI Pooling module. The improved network in this paper was compared with the Faster R-CNN model, which is a modification of feature extraction networks, such as ResNet50, ResNet101, and ResNet152, and the original VGG16 feature extraction network. The results show that the ResNet152 network had the best feature extraction effect among the feature extraction networks, and other optimizations were performed in this paper based on the use of ResNet152. In the precision−recall curve, the network in this paper showed the best effect. The improved algorithm presented in this paper was significantly improved compared with the original algorithm, with a detection accuracy of 97% and an average accuracy improvement of 7.8% in foreign object recognition.

Slope Failure and Landslide Detection in Huangdao District of Qingdao City Based on an Improved Faster R-CNN Model

To reduce the significant losses caused by slope failures and landslides, it is of great significance to detect and predict these disasters scientifically. This study focused on Huangdao District of Qingdao City in Shandong Province, using the improved Faster R-CNN network to detect slope failures and landslides. This study introduced a multi-scale feature enhancement module into the Faster R-CNN model. The module enhances the network’s perception of different scales of slope failures and landslides by deeply fusing high-resolution weak semantic features with low-resolution strong semantic features. Our experiments show that the improved Faster R-CNN model outperformed the traditional version, and that ResNet50 performed better than VGG16 with an AP value of 90.68%, F1 value of 0.94, recall value of 90.68%, and precision value of 98.17%. While the targets predicted by VGG16 were more dispersed and the false detection rate was higher than that of ResNet50, VGG16 was shown to have an advantage in predicting small-scale slope failures and landslides. The trained Faster R-CNN network model detected geological hazards of slope failure and landslide in Huangdao District, missing only two landslides, thereby demonstrating high detection accuracy. This method can provide an effective technical means for slope failures and landslides target detection and has practical implications.

Classification of the qilou (arcade building) using a robust image processing framework based on the Faster R-CNN with ResNet50

ABSTRACT Qilou (arcade building) is a particular type of Chinese historical architecture combined with western and eastern building elements, which plays a significant role in the history of modern Chinese architecture. However, the recognition and classification of the qilou mainly rely on manual inspection, suppressing the cultural dissemination and protection of qilou relics. In this paper, we present a new framework that adopts multiple image processing algorithms and a deep learning network to automate qilou classification. First, image dataset of the qilou is enhanced based on the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. Then, an improved Faster R-CNN with ResNet50 (Faster R-CNN-R) is deployed for qilou image recognition. A total of 760 images captured in Guangzhou were used for training, validation, and accuracy check of the proposed framework and several contrastive networks under the same conditions. Compared to other networks, the proposed framework works better than Faster R-CNN with VGG16 (Faster R-CNN-V) and FCOS. The accuracy of the proposed framework embedded with the Faster R-CNN-R, Faster R-CNN-V, and FCOS are 80.12%, 65.17%, and 66.35%, respectively. Based on digital images captured under different lighting conditions, the proposed framework can be used to classify nine different types of qilous, with high robustness.

Non-Destructive Internal Defect Detection of In-Shell Walnuts by X-ray Technology Based on Improved Faster R-CNN

The purpose of this study was to achieve non-destructive detection of the internal defects of in-shell walnuts using X-ray radiography technology based on improved Faster R-CNN network model. First, the FPN structure was added to the feature-extraction layer to extract richer image information. Then, ROI Align was used instead of ROI Pooling for eliminating the localization bias problem caused by the quantization operation. Finally, the Softer-NMS module was introduced to the final regression layer with the predicted bounding box for improving the localization accuracy of the candidate boxes. The results of the study indicated that the proposed network model can effectively identify internal defects of in-shell walnuts. Specifically, the discrimination accuracies of the in-shell sound, shriveled, and empty-shell walnuts were 96.14%, 91.72%, and 94.80%, respectively, and the highest overall accuracy was 94.22%. Compared to the original Faster R-CNN network model, the improved Faster R-CNN model achieved an increase of 5.86% in mAP and 5.65% in F1-value. Consequently, the proposed method can be applied for the in-shell walnuts with shriveled and empty-shell defects.