Cascaded Machine-Learning Technique for Debris Classification in Floor-Cleaning Robot Application

Ramalingam, Balakrishnan; Lakshmanan, Anirudh Krishna; Ilyas, Muhammad; Le, Anh Vu; Elara, Mohan Rajesh

doi:10.3390/app8122649

Cited by 43 publications

(39 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Solid food litter detected typically has a 97% or higher confidence level and stains, and liquid spillage has been recognized at 96% or higher confidence level, respectively. Further, the miss rate (Equation (7)) and false rate (Equation (8)) metric [13] are evaluated for the proposed litter detection framework. These two scenarios can be better understood by observing the Figure 9b,c,h.…”

Section: Evaluate the Table Cleanliness Inspection With Hsrmentioning

confidence: 99%

“…Trash detection [40] SVM + Scale-invariant feature transform (SIFT) 63 Table 4 shows CNN case study and Table 5 shows comparative analysis of CNN based litter detection schemes, which are implemented using various CNN based object detection schemes. Generally, the object detection framework efficiency has been assess through accuracy, precision, recall, and F-1 score, miss rate and false rate metric [13]. In literature point of view, compare our proposed method with the other litter detection frameworks is very hard, because each model uses a different CNN topology and training parameters.…”

Section: Case Study Algorithm Detection Accuracymentioning

confidence: 99%

“…However, configuring HSR for cleanliness inspection and food litter collection is a challenging task [11]. Because, the robots need optimal food litter (food scraps, stains, spillage) detection system and real-time planner algorithm to execute the cleaning and inspection task [12,13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Table Cleaning Task by Human Support Robot Using Deep Learning Technique

Yin

Apuroop

Tamilselvam

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

This work presents a table cleaning and inspection method using a Human Support Robot (HSR) which can operate in a typical food court setting. The HSR is able to perform a cleanliness inspection and also clean the food litter on the table by implementing a deep learning technique and planner framework. A lightweight Deep Convolutional Neural Network (DCNN) has been proposed to recognize the food litter on top of the table. In addition, the planner framework was proposed to HSR for accomplishing the table cleaning task which generates the cleaning path according to the detection of food litter and then the cleaning action is carried out. The effectiveness of the food litter detection module is verified with the cleanliness inspection task using Toyota HSR, and its detection results are verified with standard quality metrics. The experimental results show that the food litter detection module achieves an average of 96% detection accuracy, which is more suitable for deploying the HSR robots for performing the cleanliness inspection and also helps to select the different cleaning modes. Further, the planner part has been tested through the table cleaning tasks. The experimental results show that the planner generated the cleaning path in real time and its generated path is optimal which reduces the cleaning time by grouping based cleaning action for removing the food litters from the table.Sensors 2020, 20, 1698 2 of 20 vision-based techniques are widely used in cleaning robots for recognizing the litter and compute the cleaning action [14][15][16][17][18][19]. Andersen et al., built up a visual cleaning map for cleaning robots using a vision algorithm and a powerful light-transmitting diode. The sensor recognizes the grimy region and generates the dirt map by examining the surface pictures pixel-by-pixel utilizing the multi-variable statistical method [15]. David et al., proposed high-level manipulation actions for cleaning dirt from table surfaces using REEM a humanoid service robot. The author uses a background subtraction algorithm for recognizing the dirt from the table and Noisy Indeterministic Deictic (NID) rules-based learning algorithm to generate the sequence of cleaning action [16]. Ariyan et al., developed a planning algorithm for the removal of stains from non-planar surfaces where the author uses a depth-first branch-and-bound search to generate cleaning trajectories with the K-means clustering algorithm [17]. Hass et al., demonstrated the use of unsupervised clustering algorithm and Markov Decision Problem (MDP) for performing the cleaning task where unsupervised clustering algorithm is used to distinguish the dirt from surface and MDP algorithm is used to generate the maps, and transition model from clustered image is used to describe the robot cleaning action [18]. Nonetheless, these approaches have some practical issues and disadvantages for using in food court table cleaning; the detection ratio relies heavily on the textured surfaces, which makes it challenging to identify the litter type as solid...

show abstract

Section: Evaluate the Table Cleanliness Inspection With Hsrmentioning

confidence: 99%

Section: Case Study Algorithm Detection Accuracymentioning

confidence: 99%

See 1 more Smart Citation

Table Cleaning Task by Human Support Robot Using Deep Learning Technique

Yin

Apuroop

Tamilselvam

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, for various defects, thresholds often used in these algorithms need to be adjusted or it may even be necessary to redesign the algorithms [28]. CNN-based algorithms have been successfully implemented in defect detection and inspection applications including surface crack and defect detection [26-28, 30, 32], solar panel inspection [33], and cleaning inspection [34]. Cha and Choi proposed the use of CNNs [32] and Faster RCNN method [29] for better crack detection in concrete and metal surfaces.…”

Section: Introductionmentioning

confidence: 99%

Visual Inspection of the Aircraft Surface Using a Teleoperated Reconfigurable Climbing Robot and Enhanced Deep Learning Technique

Ramalingam

Vega-Heredia

Elara

et al. 2019

International Journal of Aerospace Engineering

Self Cite

View full text Add to dashboard Cite

Aircraft surface inspection includes detecting surface defects caused by corrosion and cracks and stains from the oil spill, grease, dirt sediments, etc. In the conventional aircraft surface inspection process, human visual inspection is performed which is time-consuming and inefficient whereas robots with onboard vision systems can inspect the aircraft skin safely, quickly, and accurately. This work proposes an aircraft surface defect and stain detection model using a reconfigurable climbing robot and an enhanced deep learning algorithm. A reconfigurable, teleoperated robot, named as “Kiropter,” is designed to capture the aircraft surface images with an onboard RGB camera. An enhanced SSD MobileNet framework is proposed for stain and defect detection from these images. A Self-filtering-based periodic pattern detection filter has been included in the SSD MobileNet deep learning framework to achieve the enhanced detection of the stains and defects on the aircraft skin images. The model has been tested with real aircraft surface images acquired from a Boeing 737 and a compact aircraft’s surface using the teleoperated robot. The experimental results prove that the enhanced SSD MobileNet framework achieves improved detection accuracy of aircraft surface defects and stains as compared to the conventional models.

show abstract

“…In addition, Refs. [16,17] utilize the relations between objects, and [18][19][20] utilize a cascaded network to improve detection performance.…”

Section: Introductionmentioning

confidence: 99%

Joint Pedestrian and Body Part Detection via Semantic Relationship Learning

Lan

Chen

et al. 2019

Applied Sciences

View full text Add to dashboard Cite

While remarkable progress has been made to pedestrian detection in recent years, robust pedestrian detection in the wild e.g., under surveillance scenarios with occlusions, remains a challenging problem. In this paper, we present a novel approach for joint pedestrian and body part detection via semantic relationship learning under unconstrained scenarios. Specifically, we propose a Body Part Indexed Feature (BPIF) representation to encode the semantic relationship between individual body parts (i.e., head, head-shoulder, upper body, and whole body) and highlight per body part features, providing robustness against partial occlusions to the whole body. We also propose an Adaptive Joint Non-Maximum Suppression (AJ-NMS) to replace the original NMS algorithm widely used in object detection, leading to higher precision and recall for detecting overlapped pedestrians. Experimental results on the public-domain CUHK-SYSU Person Search Dataset show that the proposed approach outperforms the state-of-the-art methods for joint pedestrian and body part detection in the wild.

show abstract

Cascaded Machine-Learning Technique for Debris Classification in Floor-Cleaning Robot Application

Cited by 43 publications

References 34 publications

Table Cleaning Task by Human Support Robot Using Deep Learning Technique

Table Cleaning Task by Human Support Robot Using Deep Learning Technique

Visual Inspection of the Aircraft Surface Using a Teleoperated Reconfigurable Climbing Robot and Enhanced Deep Learning Technique

Joint Pedestrian and Body Part Detection via Semantic Relationship Learning

Contact Info

Product

Resources

About