International Journal of Pattern Recognition and Artificial Intelligence

Object detection on hardware platforms plays a very significant role in developing driver assistance systems (DASs) with limited computational resources. Object detection for DAS is a multiclass detection problem that involves detecting various objects like cars, auto, traffic lights, bicycles, pedestrians, etc. DAS also requires accuracy, speed, and sensitivity for detecting these objects in various challenging conditions. The lighting and weather conditions pose a serious challenge for accurate object detection for DAS. This paper proposes a speed-efficient and lightweight fully convolutional neural network (CNN) architecture for object detection in adverse rainy conditions. The proposed architecture uses a CNN-based deraining network with a custom SSIM loss function in the object detection pipeline, which can give an accurate performance using limited computational and memory resources. The object detection architecture contains some architectural modifications to the existing single shot multibox detector (SSD) architecture to make it more hardware efficient and improve accuracy on small objects. It uses a trainable color transformation module using $1\times 1$ convolutions for handling the adverse lighting conditions encountered in DAS. The architecture uses feature fusion and the dilated convolution approach to enhance the accuracy of the proposed architecture on small objects. The datasets available for object detection in DAS are very imbalanced with cars as a predominant object. The class weight penalization technique is used to improve the performance of the architecture on scarcely present objects. The performance of the architecture is evaluated on well-known datasets like Kitti, Udacity, Indian Driving Dataset (IDD), and DAWN. The architecture achieves satisfactory performance in terms of mean average precision (mAP) and detection time on all these datasets. It requires three times fewer hardware resources compared to existing architectures. The lightweight nature of the proposed architecture and modification of CNN architecture with TensorRT allow the efficient implementation on the jetson nanohardware platform for prototyping, which can be integrated with other intelligent transportation systems.

Accurate identification of weed species is the premise for controlling weeds in field. But it is a challenging task due to the complexity and high-dimensional nonlinearity of the weed images in natural field. Convolutional neural networks (CNNs) model has been widely applied to image identification, but most of the CNNs models have the problems of large parameters, low identification accuracy, and single feature scale. This paper presents a novel deep neural network structure, named as MPF-Net for weed species identification. In MPF-Net, firstly, the weed images is sent into two different scales of depthwise separable convolution layers; secondly, the parallel output feature information is cross-fused, and uses the residual learning structure to increase the network model depth and feature extraction ability; finally the lightweight model PL-Model and the scale reduction module SR-Model are stacked together to construct the lightweight network. We have performed extensive experiments on real weed datasets, and compared the proposed MPF-Net against several variations of lightweight networks. The experimental results on the weed image dataset show that the proposed method is effective and feasible for weed species identification.

Visible–infrared person re-identification (VI-ReID) is a current focused area in the field of re-identification. In order to reduce the gap between two modalities in VI-ReID and improve recognition accuracy, this paper proposes a four-stream network and nonsignificant feature learning (FS-NSF) method for VI-ReID. First, the dual-intermediate modality images of visible and infrared modalities are generated by two lightweight networks, and the labels are inherited from the visible and infrared images. Second, the ResNet50 backbone network is split in order to reconstruct the network adapted to shared feature learning of the four modalities. Finally, a multi-branch, multi-scale and multi-granularity feature extraction strategy is used to extract both significant and nonsignificant features. The comparison experiments are conducted on SYSU-MM01 dataset and RegDB dataset. The experimental results show that, compared with state-of-the-arts, our method has excellent performance on both datasets, especially on the SYSU-MM01 dataset, with an increase in performance of 1.9–6.28% for each index.

Tuna fish is a popular food because of its nutritional value and taste. Demand for various species of tuna increases over time, necessitating the development of a system to sort tuna fish into distinct species in export sectors in order to accelerate the process. The work proposes an automated tuna classification system based on split octonion network. The images are initially preprocessed and divided into region images. Each region image is applied to a split octonion network with eleven layers. In addition, a split octonion channel attention module is presented, which is fed to the last two convolutional layers. The features from the three octonion networks are fused and applied to a series of dense layers. In the last layer, a softmax classifier is utilized for final classification. Results show that the proposed region-based split octonion network with attention module gives an accuracy of 98.01% on tuna database. The region-based tuna classification model is fine-tuned for the categorization of six species from QUT-FishBase dataset and Fish-Pak dataset. The system shows accuracies of 97.83% and 98.17% on QUT-FishBase and Fish-Pak datasets, respectively. The proposed methodology is also compared with existing approaches using a variety of evaluation criteria.

Automatic detection of tumors is important to speed up treatment and to increase the survival rate of patients. In brain tumor detection, Magnetic Resonance Imaging (MRI) is considered an effective imaging model, which offers the internal structure of the brain. Change detection by pre-operative as well as post-operative multimodal images is an important research area in recent decades. Thus, this paper designs a hybrid optimization algorithm-based deep learning classifier to find the percentage of change detection in multimodal images. Initially, preprocessing is progressed to eradicate the noise from MRI images and then segmentation is performed using the modified DeepJoint model. After that, the pre-operative and the post-operative MRI images are engaged for the classification of a tumor. The classification of brain tumors is performed by Deep Convolutional Neural Network (Deep CNN) trained by a Tunicate Exponential Weighted Moving Average (TEWMA) algorithm, which is the integration of Tunicate Swarm Algorithm (TSA) and Exponential Weighted Moving Average (EWMA). After classification, the volume difference and the percentage of change detection are computed by GAN trained by PS-TEWMA, which is the integration of Particle Swarm Optimization (PSO) with TSA and EWMA. The proposed PS-TEWMA-based GAN obtained lower MSE and RMSE of 0.0881 and 0.2968 by measuring the volume detection. Also, it obtained minimal MSE and RMSE of 0.102 and 0.3194 concerning the percentage of change detection.

Lane detection is one of the difficulties in implementing an advanced driving assistance system. In this paper, we show that the existing single frame-based algorithm suffers from the problem of unsatisfied detection result, which is directly caused by some extremely poor road environments (such as severe shadow occlusion and severe fading). To solve the problem, this paper proposes a multi-lane detection method based on the improved U-SegNet model. In order to get better feature extraction results, the number of network layers of the U-SegNet model is deepened, and a many-to-many structure is also proposed to improve the recognition rate of the algorithm. During the training stage, continuous multiple frames are input into the RNN module to get the feature maps for feature learning and prediction. The proposed method is tested on Caltech lane marking dataset, the results show that the proposed algorithm has good robustness and real-time performance, the multi-lane marking can be better detected under most complex road conditions, and the average detecting accuracy can achieve 96.95%.

In this paper, in view of the insufficiency of the CenterNet detector’s inability to achieve high real-time performance with high accuracy when performing object detection, we designed a new detector called A_CenterNet. In the detector, we use our newly designed lightweight Hourglass-256 model, and we also use the feature map fusion method we designed, as well as our improved attention mechanism. Through the experimental results on multiple datasets, it can be known that the A_CenterNet proposed in this paper has a competitive advantage compared with some existing classic detectors. A_CenterNet achieves the best speed-accuracy trade-off on the MS COCO dataset, with 44.6 AP at 36 FPS. Compared with CenterNet, A_CenterNet greatly improves the detection speed without loss of detection accuracy.

Body pose analysis is an important factor of human action recognition. Recently, the proposed Recurrent Neural Networks (RNNs) and deep ConvNets-based methods are showing good performances in learning sequential information. Despite these good performances, RNN lacks to efficiently learn spatial relation between body parts while deep ConvNets require a huge amount of data for training. We propose a Distance-based Neural Network (DNN) for action recognition in static images. We compute effective distances between a set of body part pairs for a given image and feed to DNN to learn effective representation of complex actions. We also propose Distance-based Convolutional Neural Network (DCNN) to learn representations from 2D images. The distances are rearranged in 2D grayscale image called as a Distance Image. This 2D representation allows the network to learn specific discriminative information between adjacent pixel distance values corresponding to different body part pairs. We evaluate our method on two real-world datasets i.e. UT-Interaction and SBU Kinect Interaction. Results show that our proposed method achieves better performance compared to the state-of-the-art approaches.

Chronic kidney disease (CKD) is one of the causes of mortality in almost all countries across the globe and the notable thing is its asymptomatic nature in the early stages. This disease is characterized by the gradual loss of kidney function in an individual. Frequently chronic kidney disease is diagnosed based on the Estimated Glomerular Filtration Rate (eGFR) determined from blood and urine tests. In order to reduce the risk factors arising due to chronic kidney disease, it is essential to be diagnosed in the earlier stages itself. This work proposes an automated chronic kidney disease detection based on the textural features of the kidney using a hybrid random forest classifier from 2D ultrasound kidney images. The proposed classifier is compared with the other competing machine learning classifiers through experimenting on a dataset of 150 images and gives a better accuracy of $96.67\%$ with $100\%$ of recall and precision, thus proving it to be promising in detecting CKD noninvasively in the early stages.

Face recognition is widely used and is one of the most challenging tasks in computer vision. In recent years, many face recognition methods based on dictionary learning have been proposed. However, most methods only focus on the resolution of the original image, and the change of resolution may affect the recognition results when dealing with practical problems. Aiming at the above problems, a method of multi-resolution dictionary learning combined with sample reverse representation is proposed and applied to face recognition. First, the dictionaries associated with multiple resolution images are learnt to obtain the first representation error. Then different auxiliary samples are generated for each test sample, and a dictionary consisted of test sample, auxiliary samples, and other classes of training samples is established to sequentially represent all training samples at this resolution, and to obtain the second representation error. Finally, a weighted fusion scheme is used to obtain the ultimate classification result. Experimental results on four widely used face datasets show that the proposed method achieves better performance and is effective for resolution change.

Keyword spotting plays a crucial role in realizing voice-based user interaction on intelligent equipment terminals and service robots. In this task, it remains challenging to achieve the balance between low memory and high precision. To better satisfy this requirement, we propose an end-to-end neural architecture with sandglass residual blocks embedded with the gated channel-wise attention mechanism. The sandglass residual blocks utilize 1D separable convolutions to extract bottleneck temporal features, which can effectively drive the model to focus more on the speech segment with lower parameters. Especially, the gated attention mechanism helps the model enhance the critical speech temporal features and suppress the useless ones and further focus on the most important part of the human speech region for keyword spotting. The experimental results on Google Speech Commands Dataset show that our proposed model has an accuracy of 97.4$\%$ with only 46K parameters. Compared with the baseline method with the highest accuracy, our model parameters are decreased by 54$\%$ and accuracy is increased by 0.8$\%$. That makes us take further step in achieving the goal of low memory and high precision.

In the future manned lunar exploration mission, astronauts would work with the lunar robots, which has a high requirement for human–robot interaction (HRI). As the accuracy of gesture recognition interaction does not fulfill the requirement for human–robot joint exploration missions, we propose the DMS-SK/BLSTM-CTC hybrid network to improve the performance of HRI. For gesture recognition, considering VGG-SK has low accuracy and complex architecture, we delete the fourth convolution module, optimize the last global pooling layer, introduce dilated convolution block and multiscale convolution block in VGG-SK, and get the DMS-SK-based gesture recognition sub-network. Compared with the traditional recognition methods, the accuracy and performance of DMS-SK improve. For speech recognition, considering that Bidirectional long–short-term memory unit (BLSTM) has the advantages of processing temporal information, and the Connectionist Temporal Classification (CTC) algorithm can simplify speech data preprocessing, we use BLSTM based on CTC as the speech recognition sub-network. Finally, we combine DMS-SK with BLSTM-CTC, and propose the DMS-SK/BLSTM-CTC hybrid network as the gesture/speech hybrid network. In addition, we use 10 gestures in the American Sign Language (ASL) dataset and 10 speech commands to construct the gesture/speech hybrid dataset. Experimental results show that compared with the pure gesture or pure speech networks, the recognition accuracy of the gesture-speech hybrid network improves by 2% and 12%, respectively, its accuracy reaches 97.38%, which fulfills the requirement of astronauts for HRI.

AI-based process model analysis has attracted more and more interest. Model quality is crucial for such research. At present, inter-organizational business process (IOBP) has been widely used in the model design and development of the distributed system. Before implementing the intelligent analysis of the IOBP model, conformance as a foundation for model quality checking plays a key role because it ensures in advance that the participants can successfully interact without violating the global communication constraints imposed by the choreography. In fact, the multi-instance participant is a common requirement in IOBP. This paper provides a formal approach and framework supporting the conformance between BPMN choreography and collaboration while considering multi-instance participants and message communication modes. As a core, the formalization proposed is based on BNF syntax and structured CSP# processes. It can well support multi-instance features and multiple communication modes. Combined with CSP#, the formal definitions of communication modes and verification properties are given. On this basis, an integrated framework is provided to support automated formal verification referring to multiple communication modes. Finally, a set of experiments is conducted to demonstrate the effectiveness of the proposal.

Random access systems are potential for Internet of Things in the future wireless communication network for its operational simplicity. Irregular repetition slotted ALOHA (IRSA) system is one of the high-efficiency random access systems. In this paper, performance analysis of the irregular repetition slotted ALOHA systems with short-packet, i.e. finite-blocklength, transmission for the quasi-static Rayleigh fading channel is given. A cumulative distribution function of signal-to-interference power ratio (SIR) is derived and thus a closed-form expression of an average packet error probability (PEP) at the SIR with short packets for the Rayleigh fading channel is given. The closed-form expression makes it possible to optimize the degree distributions at a specific blocklength in the sense that the systems give the maximum system load.