Deep learning in food category recognition

Zhang, Yudong; Deng, Lijia; Zhu, Hengde; Wang, Wei; Ren, Zeyu; Zhou, Qinghua; Lu, Siyuan; Sun, Shiting; Zhu, Ziquan; Gorriz, Juan Manuel; Wang, Shuihua

doi:10.1016/j.inffus.2023.101859

Cited by 205 publications

(35 citation statements)

References 231 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed feature fusion model's robustness to data contamination (data noise) is experimentally verified. Zhang et al 33 proposed a deep learning approach for food category recognition. By effectively combining the strengths of convolutional neural networks and recurrent neural networks, the method facilitates efficient classification of food images through multi‐scale feature extraction and contextual information integration.…”

Section: Related Workmentioning

confidence: 99%

CGS‐Net: A classification‐guided framework for automated infection segmentation of COVID‐19 from CT images

Zhou,

Wang,

Wang

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Automated segmentation of lung lesions in CT images of COVID‐19 based on deep learning holds great potential for comprehending the advancement of the disease and establishing suitable treatment approaches. However, the complex background, indistinct boundaries, varying sizes and distributions of infected regions, and high similarity to other lung diseases pose substantial challenges. To address these issues, we propose a joint deep learning‐based framework, named Classification‐Guided Segmentation Network (CGS‐Net), for COVID‐19 segmentation. The framework comprises a classification and a segmentation sub‐network, which are trained sequentially. Initially, the classification sub‐network learns the feature information of COVID‐19 and other pneumonia classifications. Subsequently, the trained classification sub‐network assists in the training of the segmentation sub‐network. In addition, several key modules are employed to construct the network, including the Multi‐scale Feature Mapping Module (MSFM), the Context Information Module (Context), and the Axial Attention Fusion Module (AAFM). The MSFM employs dilated convolutions to extract multi‐scale features with an emphasis on spatial and channel information. The Context combines strip pooling and average pooling to aggregate contextual and global information. Finally, the AAFM is employed to incorporate contextual information through an axial attention mechanism. Extensive experimental results demonstrate that our proposed method outperforms competing approaches in the segmentation task.

show abstract

Section: Related Workmentioning

confidence: 99%

CGS‐Net: A classification‐guided framework for automated infection segmentation of COVID‐19 from CT images

Zhou,

Wang,

Wang

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…Deep learning and AI have revolutionized all the domains of human life, be it medical and health applications as a tool in the fight against COVID-19, 11,12 or food category recognition, for detecting food type and quantity to analyzing quality and components. 13 Both traditional ML and deep learning have proved effective in the classification task at hand.…”

Section: Review Of Past Workmentioning

confidence: 99%

An enhanced framework for identifying brain tumor using discrete wavelet transform, deep convolutional network, and feature fusion‐based machine learning techniques

Mehrotra,

Ansari,

Agrawal

et al. 2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Today, the histological study of biopsy specimens is still used to diagnose brain tumors (BTs). This existing procedure is intrusive, arduous, and liable to mistakes. These downsides highlight the standing of employing a completely computerized process for identifying the evolution of tumors in the brain. A primary BT affects an estimated 0.7 million persons in the United States now and more are expected to be detected in the coming years. The ability to categorize magnetic resonance (MR) brain images into ordinary and pathological categories has the boundless ability to significantly diminish the burden on the radiologist. Pre‐processing, extraction, and reduction of features along with their classification are the parameters of statistical‐based methodologies that have been frequently used for this purpose. In this work, an enhanced framework for the identification of the BT is proposed using discrete wavelet transform (DWT), deep convolutional network (DCN), and machine learning (ML). As DWT is primarily used for image compression and denoising applications however in the presented research work it has been utilized for extricating pivotal features from the MR images using the feature fusion technique. DCN is also utilized for the extraction of pivotal deep features which are then combined with the wavelet‐based features for the purpose of BT identification. The classification of tumorous and non‐tumorous MR images is done using ML applications. The results obtained from the proposed model exhibit an utmost accuracy of 99.5% with an area under curve of 1 in identifying tumorous and non‐tumorous MR images as compared to various state‐of‐the‐art models. The proposed model can be efficiently used for assisting radiologists and medical experts in validating their decisions for BT identification.

show abstract

“…Over the past few years, deep convolutional neural networks (DCNNs) have displayed impressive achievement in a range of computer vision tasks, such as image classification, semantic segmentation, and object detection. [ 3–6 ] In contrast to the traditional “physics‐based or manual‐based” method, DCNNs are a type of data‐driven artificial neural network that is particularly well‐suited for processing image data and has proven to be highly effective in recognizing patterns and features in images. [ 7 ] This has led to increased interest in using DCNNs for cell segmentation and counting in medical imaging, as they can automate and improve the accuracy of these tasks compared to traditional manual counting methods.…”

Section: Introductionmentioning

confidence: 99%

SSRNet: A Deep Learning Network via Spatial‐Based Super‐resolution Reconstruction for Cell Counting and Segmentation

Deng

Zhou

Wang

et al. 2023

Advanced Intelligent Systems

Self Cite

View full text Add to dashboard Cite

Cell counting and segmentation are critical tasks in biology and medicine. The traditional methods for cell counting are labor‐intensive, time‐consuming, and prone to human errors. Recently, deep learning‐based cell counting methods have become a trend, including point‐based counting methods, such as cell detection and cell density prediction, and non‐point‐based counting, such as cell number regression prediction. However, the point‐based counting method heavily relies on well‐annotated datasets, which are scarce and difficult to obtain. On the other hand, nonpoint‐based counting is less interpretable. The task of cell counting by dividing it into two subtasks is approached: cell number prediction and cell distribution prediction. To accomplish this, a deep learning network for spatial‐based super‐resolution reconstruction (SSRNet) is proposed that predicts the cell count and segments the cell distribution contour. To effectively train the model, an optimized multitask loss function (OM loss) is proposed that coordinates the training of multiple tasks. In SSRNet, a spatial‐based super‐resolution fast upsampling module (SSR‐upsampling) is proposed for feature map enhancement and one‐step upsampling, which can enlarge the deep feature map by 32 times without blurring and achieves fine‐grained detail and fast processing. SSRNet uses an optimized encoder network. Compared with the classic U‐Net, SSRNet's running memory read and write consumption is only 1/10 of that of U‐Net, and the total number of multiply and add calculations is 1/20 of that of U‐Net. Compared with the traditional sampling method, SSR‐upsampling can complete the upsampling of the entire decoder stage at one time, reducing the complexity of the network and achieving better performance. Experiments demonstrate that the method achieves state‐of‐the‐art performance in cell counting and segmentation tasks. The method achieves nonpoint‐based counting, eliminating the need for exact position annotation of each cell in the image during training. As a result, it has demonstrated excellent performance on cell counting and segmentation tasks. The code is public on GitHub (https://github.com/Roin626/SSRnet).

show abstract

Deep learning in food category recognition

Cited by 205 publications

References 231 publications

CGS‐Net: A classification‐guided framework for automated infection segmentation of COVID‐19 from CT images

CGS‐Net: A classification‐guided framework for automated infection segmentation of COVID‐19 from CT images

An enhanced framework for identifying brain tumor using discrete wavelet transform, deep convolutional network, and feature fusion‐based machine learning techniques

SSRNet: A Deep Learning Network via Spatial‐Based Super‐resolution Reconstruction for Cell Counting and Segmentation

Contact Info

Product

Resources

About