In high volume production processes such as injection molding, inline inspection of parts is generally not feasible due to additional sensor requirements that incur costs. Production facilities often inspect for defective parts at the lot-level after multiple production runs and not for each run. Individual lots are accepted or rejected based on the manufacturer’s acceptable quality levels determined by the number of faulty parts in a lot. In this paper, a Lot-level Convolutional Neural Network (L-CNN) is proposed that implements two variants to improve quality estimation at lot-level using run-level sensor data. Layer-wise L-CNN uses a separation of layers within the model architecture to extract relevant features for each run. Custom loss L-CNN utilizes a standard CNN architecture with a custom loss function to handle the data’s multi-input-single-output structure. The model is evaluated using data from injection molding and wafer testing applications in the semiconductor industry. L-CNN achieves better F1and G-Mean scores compared with existing benchmarks. Even though trained with partial information (i.e., lot-level quality), custom loss L-CNN provides both lot-level and run-level quality predictions.

Factories are increasingly pushing towards automation and data-centric approaches under the current Industry 4.0 standards. Early-stage product quality estimation is identified as one of the solutions that can significantly reduce manufacturing cost and wastage. However, quality estimation is a classification problem that is inherently challenging for traditional data-driven algorithms due to its imbalanced nature. In this paper, a framework is proposed, that combines the feature extraction capabilities of convolutional neural networks and the domain knowledge characteristics of fuzzy systems. The proposed method addresses data imbalance using an ensemble of class-biased individuals, that learn features using a class-weighted loss function. Experiments were conducted using a benchmark dataset and production data acquired from the semiconductor industry. Improvements were noted for G-Mean and ROC-AVC values when compared to existing algorithms.

It has been shown that developing nano-satellites is a good platform to motivate and educate students in Science, Technology, Engineering and Mathematics (STEM) disciplines. In this paper, the authors have explained an innovative project management method, called major-project-review, and its implementations to supervise different batches of multidisciplinary final year engineering students to develop a CubeSat as their Final Year Projects (FYPs) in three to four years. The major-project-review approach has been successfully verified in our first CubeSat project, Galassia, and is continuously used to develop our second CubeSat, Galassia2. From the success of Galassia and student feedbacks, the method has been proven to be very efficient to control and monitor the CubeSat development progress, and to motivate and educate engineering students. With our continuous implementations in Galassia2 project, we hope to collect more experiences and data to further optimize the approach.

An elastic graph matching algorithm using biologically inspired features is proposed for the recognition of hand postures. Each node in the graph is labeled using an image feature extracted using the computational model of the ventral stream of visual cortex. The graph nodes are assigned to geometrically significant positions in the hand image, and, the model graphs are created. Bunch graph method is used for modeling the variability in hand posture appearance. Recognition of a hand posture is done by the elastic graph matching between the model graphs and the input image. A radial basis function is used as the similarity function for the matching process. The proposed algorithm is tested on a 10 class hand posture database which consists of 478 grey scale images with light and dark backgrounds. The algorithm provided better recognition accuracy (96.35%) compared to the reported results (93.77%) in the literature.

In this paper, an image is regarded as a collection of image patches that can be referred to as points with certain intrinsic structures/patterns in high-dimensional space. These structures contain vital information of image features and thus provide a novel method for image feature representation. To discover these intrinsic structures, we first propose neighborhood linear embedding (NLE), an unsupervised learning algorithm, to discover neighborhood relationship and global distribution of input data simultaneously. Secondly, NLE is extended to discover the clustering structure of data by incorporating with a Euclidean distance histogram and a series of band pass filters. Finally, by combining with a dimensionality reduction technique, the discovered intrinsic structures are visualized and manipulated in low-dimensional space in the format known as embeddings. The proposed NLE allows the discovery process to adapt to the characteristics of input data. In addition, it is revealed that an image feature composed of image patches can be tracked by tracking the contour containing embeddings of the corresponding image patches

In this paper, motion estimation is proposed by fusing audio and video sensor data. The audio system consists of three microphones arranged on a Y-shaped structure, mounted on a pan-tilt camera. The camera forms the video system. Together, the audio and video system enables the 3D position of the sound source to be estimated. Using the position estimates, a motion model, consisting of the translational velocity and acceleration of the source, is in turn estimated using a Kalman filter. The motion model allows the sound source to be tracked in real time. This fusion estimation system has many potential applications such as video conferencing and security monitoring for intruders. Simulation results show that the motion estimation is satisfactory.

In this paper, relative azimuth audio localization is investigated using two microphones and an asymmetric mask. The Helmholtz-Kirchhoff integral method is employed to obtain information on ITD (Interaural Time Differences) and IID (Interaural Intensity Differences) which are two important cues for sound localization. It is proved that the proposed system can locate a sound source with relative azimuth in nearly 360 degrees.

Utilises computational intelligence techniques, such as genetic algorithms (GAs) and multi-objective evolutionary algorithms (MOEAs), to design a Wiener-model controller for regulating the pH level in an acid-base titration process. A Wiener-model control structure comprises (i) an inverse model of the system nonlinearity and (ii) a simple linear controller. The inverse model serves to simplify the control problem by eliminating the bulk of the nonlinear characteristics from the pH control loop. A PID controller can then be used to control the linearised system. A GA was employed to identify the parameters of the inverse titration equation while the PID parameters were obtained by using a MOEA. Experimental results demonstrating the viability of the proposed methodology are presented.