Jinxia Shang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jinxia Shang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Xiaowu Li
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, 727 South Jingming Road, Chenggong District, Kunming 650500, China
6 shared publications Shiyi Tian
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China
2 shared publications- Zhichao Wang
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, 727 South Jingming Road, Chenggong District, Kunming 650500, China
2 shared publications - Fan Li
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China
1 shared publication - Gang Chen
Kunming University of Science and Technology
1 shared publication - Haozhe Tang
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, Yunnan, China
1 shared publication
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (22) , 55366-55381
Hierarchical Adaptive Energy-Efficient Protocol for LoRa-Mesh Networks: A Data Fusion and Wake-on-Air-Based Approach
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (22) , 55366-55381
Hierarchical Adaptive Energy-Efficient Protocol for LoRa-Mesh Networks: A Data Fusion and Wake-on-Air-Based Approach
- Open Access
- 1 September 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (9) , 412
A Text-Guided Lesion Mining Vision–Language Ordinal Classification Framework for Diabetic Retinopathy Grading
Diabetic retinopathy (DR) is a major retinal disease that can cause visual impairment and irreversible blindness. Accurate automated DR grading is essential for large-scale screening and timely clinical intervention. However, most existing methods rely primarily on visual features for classification. Moreover, they often overlook the ordinal structure of DR severity and the intra-class phenotypic heterogeneity arising from diverse lesion combinations. To address these issues, based on the semantic prior information provided by RetiZero, we propose a text-guided lesion mining vision–language ordinal classification framework for DR grading. The proposed framework introduces a text-guided cross-layer lesion mining module that exploits semantic response differences between normal-tissue and lesion-related textual prompts, thereby guiding multi-level visual patch features toward lesion regions relevant to DR grading. To explicitly model the ordered progression of DR severity, we design a conditional ordinal regression branch and an ordinal distribution alignment strategy that jointly encourage the predictions to follow the inherent order of DR grades. Moreover, we introduce a multi-center feature constraint to capture diverse intra-grade phenotypic patterns and enhance feature discriminability. Experiments on APTOS 2019 show that the proposed method achieves 86.3% accuracy, 90.6% AUC, and 70.9% Macro-F1, which improved by 2.4, 0.7, and 5.5 percentage points compared to RetiZero. Furthermore, under the standardized leave-one-domain-out protocol of GDRNet, the proposed method achieves the highest reported average accuracy of 58.6% across six public DR datasets, exceeding the reported result of PAF (54.6%) by 4.0 percentage points. Nevertheless, our approach is limited in F1 and AUC metrics, for which GDRNet delivers superior performance. These results suggest that the proposed framework can improve DR grading performance and the cross-dataset generalization ability of the model to a certain extent.
- 1 September 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (9) , 412
Open AccessA Text-Guided Lesion Mining Vision–Language Ordinal Classification Framework for Diabetic Retinopathy Grading
Diabetic retinopathy (DR) is a major retinal disease that can cause visual impairment and irreversible blindness. Accurate automated DR grading is essential for large-scale screening and timely clinical intervention. However, most existing methods rely primarily on visual features for classification. Moreover, they often overlook the ordinal structure of DR severity and the intra-class phenotypic heterogeneity arising from diverse lesion combinations. To address these issues, based on the semantic prior information provided by RetiZero, we propose a text-guided lesion mining vision–language ordinal classification framework for DR grading. The proposed framework introduces a text-guided cross-layer lesion mining module that exploits semantic response differences between normal-tissue and lesion-related textual prompts, thereby guiding multi-level visual patch features toward lesion regions relevant to DR grading. To explicitly model the ordered progression of DR severity, we design a conditional ordinal regression branch and an ordinal distribution alignment strategy that jointly encourage the predictions to follow the inherent order of DR grades. Moreover, we introduce a multi-center feature constraint to capture diverse intra-grade phenotypic patterns and enhance feature discriminability. Experiments on APTOS 2019 show that the proposed method achieves 86.3% accuracy, 90.6% AUC, and 70.9% Macro-F1, which improved by 2.4, 0.7, and 5.5 percentage points compared to RetiZero. Furthermore, under the standardized leave-one-domain-out protocol of GDRNet, the proposed method achieves the highest reported average accuracy of 58.6% across six public DR datasets, exceeding the reported result of PAF (54.6%) by 4.0 percentage points. Nevertheless, our approach is limited in F1 and AUC metrics, for which GDRNet delivers superior performance. These results suggest that the proposed framework can improve DR grading performance and the cross-dataset generalization ability of the model to a certain extent.
- Open Access
- 9 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (14) , 3016
Identity-Aware Dynamic Indoor Passage Monitoring Using RFID Tag Arrays and Distance-Aware RSSI Temporal Modeling
Indoor passage monitoring requires people-count estimation, behavior recognition and identity association while avoiding privacy-invasive sensing. This paper reports an RFID-only corridor-style passage-monitoring framework that combines wearable EPC identity tags, a wall-mounted passive tag array and distance-aware RSSI temporal modeling. Identity-tag readings and multi-channel tag-array RSSI sequences are aligned on a unified reader-timestamp timeline. An effective behavior duration (EBD) mechanism is used as a rule-based valid-window screening step before model inference. Target count, behavior state and identity association are then estimated using distance-aware LSTM temporal fusion. The RSSI attenuation relationship is used as a physical motivation for feature design and distance-mode interpretation, rather than as an explicit loss constraint or separately validated physical prior. A subject-independent dataset was collected with 18 array tags and 18 volunteers across single-person, multi-person, crossing and following scenarios, yielding 11,148 EBD-valid windows. Under the EBD-valid event-window protocol, the proposed method achieves a mean absolute error (MAE) of 0.226, a root mean square error (RMSE) of 0.357, a Macro-F1 of 0.934 and an identity-association accuracy of 0.961. Relative to threshold rules, traditional machine learning and a single-branch LSTM evaluated on the same dataset, the MAE is reduced by 59.5%, 46.1% and 28.9%, respectively, and the RMSE is reduced by 55.8%, 43.5% and 28.9%, respectively. The results indicate that native RFID identity, spatial tag-array sensing, EBD screening and distance-aware temporal modeling can jointly support privacy-preserving passage monitoring under controlled corridor-like conditions.
- 9 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (14) , 3016
Open AccessIdentity-Aware Dynamic Indoor Passage Monitoring Using RFID Tag Arrays and Distance-Aware RSSI Temporal Modeling
Indoor passage monitoring requires people-count estimation, behavior recognition and identity association while avoiding privacy-invasive sensing. This paper reports an RFID-only corridor-style passage-monitoring framework that combines wearable EPC identity tags, a wall-mounted passive tag array and distance-aware RSSI temporal modeling. Identity-tag readings and multi-channel tag-array RSSI sequences are aligned on a unified reader-timestamp timeline. An effective behavior duration (EBD) mechanism is used as a rule-based valid-window screening step before model inference. Target count, behavior state and identity association are then estimated using distance-aware LSTM temporal fusion. The RSSI attenuation relationship is used as a physical motivation for feature design and distance-mode interpretation, rather than as an explicit loss constraint or separately validated physical prior. A subject-independent dataset was collected with 18 array tags and 18 volunteers across single-person, multi-person, crossing and following scenarios, yielding 11,148 EBD-valid windows. Under the EBD-valid event-window protocol, the proposed method achieves a mean absolute error (MAE) of 0.226, a root mean square error (RMSE) of 0.357, a Macro-F1 of 0.934 and an identity-association accuracy of 0.961. Relative to threshold rules, traditional machine learning and a single-branch LSTM evaluated on the same dataset, the MAE is reduced by 59.5%, 46.1% and 28.9%, respectively, and the RMSE is reduced by 55.8%, 43.5% and 28.9%, respectively. The results indicate that native RFID identity, spatial tag-array sensing, EBD screening and distance-aware temporal modeling can jointly support privacy-preserving passage monitoring under controlled corridor-like conditions.
- Open Access
- 17 June 2026
- Published by MDPI AG in Preprints.org
Identity-Aware Dynamic Indoor Passage Monitoring Using RFID Tag Arrays and Distance-Aware RSSI Temporal Modeling
Indoor passage monitoring requires people-count estimation, behavior recognition and identity association while avoiding privacy-invasive sensing. This paper presents an identity-aware RFID sensing framework that combines wearable EPC identity tags, a wall-mounted passive tag array and distance-aware RSSI temporal modeling. Identity-tag readings and multi-channel tag-array RSSI sequences are aligned on a unified reader-timestamp timeline. An effective behavior duration (EBD) mechanism removes invalid or weak-occlusion windows before model inference. Target count, behavior state and identity association are then estimated using a three-branch LSTM constrained by physical RSSI attenuation priors. A subject-independent dataset was collected with 18 array tags and 18 volunteers across single-person, multi-person, crossing and following scenarios, yielding 11,148 EBD-valid windows. On the test set, the proposed method achieves a mean absolute error (MAE) of 0.226, a root mean square error (RMSE) of 0.357, a Macro-F1 of 0.934 and an identity-association accuracy of 0.961. Relative to threshold rules, traditional machine learning and a single-branch LSTM, the MAE is reduced by 59.5%, 46.1% and 28.9%, respectively, and the RMSE is reduced by 55.8%, 43.5% and 28.9%, respectively. The results show that native RFID identity, spatial tag-array sensing, EBD screening and distance-aware temporal modeling can jointly support low-cost, reproducible and privacy-preserving people-flow monitoring in smart-building environments.
- 17 June 2026
- Published by MDPI AG in Preprints.org
Open AccessIdentity-Aware Dynamic Indoor Passage Monitoring Using RFID Tag Arrays and Distance-Aware RSSI Temporal Modeling
Indoor passage monitoring requires people-count estimation, behavior recognition and identity association while avoiding privacy-invasive sensing. This paper presents an identity-aware RFID sensing framework that combines wearable EPC identity tags, a wall-mounted passive tag array and distance-aware RSSI temporal modeling. Identity-tag readings and multi-channel tag-array RSSI sequences are aligned on a unified reader-timestamp timeline. An effective behavior duration (EBD) mechanism removes invalid or weak-occlusion windows before model inference. Target count, behavior state and identity association are then estimated using a three-branch LSTM constrained by physical RSSI attenuation priors. A subject-independent dataset was collected with 18 array tags and 18 volunteers across single-person, multi-person, crossing and following scenarios, yielding 11,148 EBD-valid windows. On the test set, the proposed method achieves a mean absolute error (MAE) of 0.226, a root mean square error (RMSE) of 0.357, a Macro-F1 of 0.934 and an identity-association accuracy of 0.961. Relative to threshold rules, traditional machine learning and a single-branch LSTM, the MAE is reduced by 59.5%, 46.1% and 28.9%, respectively, and the RMSE is reduced by 55.8%, 43.5% and 28.9%, respectively. The results show that native RFID identity, spatial tag-array sensing, EBD screening and distance-aware temporal modeling can jointly support low-cost, reproducible and privacy-preserving people-flow monitoring in smart-building environments.
- Open Access
Data-Driven Energy-Saving Methods Based on LoRa-Mesh Hierarchical Network
As a reliable and high-potential wireless communication technology for the Internet of Things (IoT), LoRa excels in long-distance and low-power transmission. The star topology adopted by traditional LoRaWAN suffers from poor deployment flexibility and insufficient scalability in scenarios with complex terrain or harsh environments. LoRa-Mesh networks can effectively solve coverage challenges through characteristics such as multi-hop and self-organization; however, the relay and forwarding requirements of nodes also introduce new challenges in energy consumption management. To address the energy consumption management challenges of LoRa-Mesh, this paper proposes a Data-Driven Energy Saving (DDES) protocol. It flexibly sets and dynamically fine-tunes node sleep durations based on data changes, constructs an efficient energy-saving framework through uplink data streams, and implements precise control over nodes via downlink post-analysis messages to achieve on-demand energy saving. Simulation results in the smart agriculture scenario of soil moisture monitoring and irrigation show that compared with protocols without a sleep mechanism, the battery life of the LoRa-Mesh network using the DDES protocol is extended by approximately 20 times. The proposed protocol breaks through the limitations of fixed sleep schemes, realizes refined and flexible division of sleep regions, and exhibits significant advantages in LoRa network energy saving.
Open AccessData-Driven Energy-Saving Methods Based on LoRa-Mesh Hierarchical Network
As a reliable and high-potential wireless communication technology for the Internet of Things (IoT), LoRa excels in long-distance and low-power transmission. The star topology adopted by traditional LoRaWAN suffers from poor deployment flexibility and insufficient scalability in scenarios with complex terrain or harsh environments. LoRa-Mesh networks can effectively solve coverage challenges through characteristics such as multi-hop and self-organization; however, the relay and forwarding requirements of nodes also introduce new challenges in energy consumption management. To address the energy consumption management challenges of LoRa-Mesh, this paper proposes a Data-Driven Energy Saving (DDES) protocol. It flexibly sets and dynamically fine-tunes node sleep durations based on data changes, constructs an efficient energy-saving framework through uplink data streams, and implements precise control over nodes via downlink post-analysis messages to achieve on-demand energy saving. Simulation results in the smart agriculture scenario of soil moisture monitoring and irrigation show that compared with protocols without a sleep mechanism, the battery life of the LoRa-Mesh network using the DDES protocol is extended by approximately 20 times. The proposed protocol breaks through the limitations of fixed sleep schemes, realizes refined and flexible division of sleep regions, and exhibits significant advantages in LoRa network energy saving.
- Open Access
- 23 January 2026
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 12 (3)
Semcm: semantic enhancement and multi-center modeling for maize leaf disease recognition
- 23 January 2026
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 12 (3)
Open AccessSemcm: semantic enhancement and multi-center modeling for maize leaf disease recognition
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 1
FSL-DGNet: A Lightweight Differential Gradient Network for RFID Human Activity Recognition
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 1
Open AccessFSL-DGNet: A Lightweight Differential Gradient Network for RFID Human Activity Recognition
- Open Access
- 22 October 2025
- Published by MDPI AG in Electronics
- Vol. 14 (21) , 4135
Forest Fire Monitoring and Energy Optimization Based on LoRa-Mesh Wireless Communication Technology
Forest fire monitoring is of great significance for ecological protection and public safety. This study proposes a monitoring technology based on LoRa-Mesh (Long Range-Mesh) wireless communication, integrating temperature and humidity sensing, image acquisition, fire identification, data transmission, and energy-saving optimization. To address the limitations of traditional LoRa networks in flexibility and energy consumption, a Layered Dynamic Synchronization Energy-saving (LDSE) protocol is designed. By constructing a hierarchical network, employing implicit route exploration, multi-channel and multi-path communication, and time synchronization optimization, the protocol significantly reduces packet loss rate and system energy consumption. Experimental results demonstrate that the LDSE protocol outperforms the traditional Ad hoc On-Demand Distance Vector Routing Protocol (AODV) in terms of packet loss rate, energy consumption, and latency. Additionally, the proposed energy-saving algorithm significantly reduces system power consumption, with the node sleep-relay mode exhibiting optimal energy efficiency. Experimental verification confirms that the system achieves high reliability, low power consumption, and efficient data transmission, providing an effective IoT solution for forest fire prevention.
- 22 October 2025
- Published by MDPI AG in Electronics
- Vol. 14 (21) , 4135
Open AccessForest Fire Monitoring and Energy Optimization Based on LoRa-Mesh Wireless Communication Technology
Forest fire monitoring is of great significance for ecological protection and public safety. This study proposes a monitoring technology based on LoRa-Mesh (Long Range-Mesh) wireless communication, integrating temperature and humidity sensing, image acquisition, fire identification, data transmission, and energy-saving optimization. To address the limitations of traditional LoRa networks in flexibility and energy consumption, a Layered Dynamic Synchronization Energy-saving (LDSE) protocol is designed. By constructing a hierarchical network, employing implicit route exploration, multi-channel and multi-path communication, and time synchronization optimization, the protocol significantly reduces packet loss rate and system energy consumption. Experimental results demonstrate that the LDSE protocol outperforms the traditional Ad hoc On-Demand Distance Vector Routing Protocol (AODV) in terms of packet loss rate, energy consumption, and latency. Additionally, the proposed energy-saving algorithm significantly reduces system power consumption, with the node sleep-relay mode exhibiting optimal energy efficiency. Experimental verification confirms that the system achieves high reliability, low power consumption, and efficient data transmission, providing an effective IoT solution for forest fire prevention.
- Open Access
- 4 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 101592-101608
Research on Citrus Leaf Disease Recognition Using Class-Agnostic Contrastive Learning and Supervised Organizational Mapping
Early detection of plant diseases is crucial for enhancing agricultural productivity and ensuring crop protection. While computer vision offers scalable alternatives to manual inspection, existing methods face two understudied challenges: (1) performance degradation in cluttered real-world conditions, and (2) ineffective feature separation for randomly distributed lesions with diverse symptoms. Notably, current research predominantly relies on controlled-environment datasets, lacking publicly accessible benchmarks with authentic field complexities. To address these, we present the first large-scale citrus leaf disease dataset (5,763 images, 6 categories) capturing complex field conditions including varying illumination, occlusion, multi-scale lesions, and deceptive backgrounds. Our key methodological innovations include a three-stage weakly supervised framework that systematically resolves background interference and feature redundancy: (i) Class-agnostic contrastive localization enabling precise lesion mapping without region annotations, (ii) Stable feature initialization through region-guided classification training, (iii) Sparse orthogonal mapping with experimentally-determined nodes introducing orthogonality constraints to enhance feature separation. Through feature alignment and feature fusion, an accuracy of 97.1% is achieved (+2.4% vs ResNet-18). Evaluations on citrus leaf and public datasets demonstrate superior accuracy and interpretable lesion localization, advancing plant disease diagnosis.
- 4 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 101592-101608
Open AccessResearch on Citrus Leaf Disease Recognition Using Class-Agnostic Contrastive Learning and Supervised Organizational Mapping
Early detection of plant diseases is crucial for enhancing agricultural productivity and ensuring crop protection. While computer vision offers scalable alternatives to manual inspection, existing methods face two understudied challenges: (1) performance degradation in cluttered real-world conditions, and (2) ineffective feature separation for randomly distributed lesions with diverse symptoms. Notably, current research predominantly relies on controlled-environment datasets, lacking publicly accessible benchmarks with authentic field complexities. To address these, we present the first large-scale citrus leaf disease dataset (5,763 images, 6 categories) capturing complex field conditions including varying illumination, occlusion, multi-scale lesions, and deceptive backgrounds. Our key methodological innovations include a three-stage weakly supervised framework that systematically resolves background interference and feature redundancy: (i) Class-agnostic contrastive localization enabling precise lesion mapping without region annotations, (ii) Stable feature initialization through region-guided classification training, (iii) Sparse orthogonal mapping with experimentally-determined nodes introducing orthogonality constraints to enhance feature separation. Through feature alignment and feature fusion, an accuracy of 97.1% is achieved (+2.4% vs ResNet-18). Evaluations on citrus leaf and public datasets demonstrate superior accuracy and interpretable lesion localization, advancing plant disease diagnosis.
- 28 December 2023
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (01) , 013001-15
Salient object detection based on hierarchical segmentation and objectness-guided
Unsupervised salient object detection aims to automatically detect important or attractive target objects in the scene without any user annotation. Compared to fully supervised, it can save a lot of manpower and resources invested in pixel-level annotation. However, the performance of existing unsupervised methods is still far from satisfactory because of the interference of complex backgrounds and the lack of knowledge in the semantic layer of scene objects. To solve these challenging problems, in this paper, we propose a new bottom-up method to better highlight the integral salient object region by exploiting the co-occurrence of saliency and objectness properties. Specifically, saliency estimation effectively utilizes low-level visual features based on high-quality hierarchical image segmentation (convolutional oriented boundaries). The objectness scoring is mainly obtained through the improved object proposals mapping map. Experiment results demonstrate that the proposed method can efficiently remove the interference of non-salient objects and complex background while preserving more salient object regions.
- 28 December 2023
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (01) , 013001-15
Salient object detection based on hierarchical segmentation and objectness-guided
Unsupervised salient object detection aims to automatically detect important or attractive target objects in the scene without any user annotation. Compared to fully supervised, it can save a lot of manpower and resources invested in pixel-level annotation. However, the performance of existing unsupervised methods is still far from satisfactory because of the interference of complex backgrounds and the lack of knowledge in the semantic layer of scene objects. To solve these challenging problems, in this paper, we propose a new bottom-up method to better highlight the integral salient object region by exploiting the co-occurrence of saliency and objectness properties. Specifically, saliency estimation effectively utilizes low-level visual features based on high-quality hierarchical image segmentation (convolutional oriented boundaries). The objectness scoring is mainly obtained through the improved object proposals mapping map. Experiment results demonstrate that the proposed method can efficiently remove the interference of non-salient objects and complex background while preserving more salient object regions.