This research presents a novel post-processing method for convolutional neural networks (CNNs) in character recognition, specifically designed to handle inconsistencies and irregularities in character shapes. Convolutional Neural Networks (CNNs) are powerful tools for recognizing and learning character shapes directly from source images, making them well-suited for recognition of characters that contain inconsistencies in their shapes. However, when applied to multi-object detection for character recognition, CNNs require post-processing to convert the recognized characters into code sequences, which has so far limited their applicability. The developed method solves this problem by directly post-processing the inconsistent characters identified by the convolutional neural model into labels corresponding to the source image. An experiment with real pharmaceutical packaging images demonstrates the functionality of the method, showing that it can handle different numbers of characters and labels effectively. As a scientific contribution to the fields of imaging and deep learning, this research opens new possibilities for future studies, particularly in the development of more accurate and efficient multi-object character recognition with post-processing and their application to new areas.
We present several stereoscopic 3D radiographs, obtained in a clinical setting using a technique requiring minimal operator training and no new technology. Reviewing known perceptual advantages in stereoscopic imaging, we argue the benefits for diagnosis and treatment planning primarily in orthopedics, with opportunities likely extended to rheumatology, oncology, and angiology (vascular medicine/surgery). These advantages accrue with the marginal additional cost of capturing just two or three supplementary radiographs using the proposed method. Presently, computed tomography (CT) scanning is standard for obtaining 3D imagery. We discuss relative advantages of stereoscopic 3D radiography (3DSR) in imaging resolution, cost, availability, and radiation dose. Further discussion will describe obstacles and challenges likely to be encountered in clinical implementation of 3DSR, to be mitigated through targeted training of clinicians and technicians. Further research is needed to explore and empirically validate the potential value of 3DSR. We hope to pave the way for this more accessible and cost-effective 3D imaging solution, enhancing diagnostic capabilities and treatment planning, especially in resource-constrained settings. (All 3D radiographs presented in this paper are in the red/cyan color anaglyph format. 3D anaglyph glasses are commonly available online, or in most comics bookstores. This report’s images can be found in L-R stereo-pair format, suitable for 3D viewing with a 3D screen or stereoscope, at this URL: https://www.starosta.com/3DSR/)
Sublimity has long been a theme in aesthetics, as one of the human emotions experienced when perceiving vastness, terror, or ambiguity. This study investigated the visual conditions that evoke sublimity using virtual reality (VR). Participants observed sublime content, developed based on previous research, under two factors conditions of (1) wide or narrow fields of view (FOV), and (2) 2D or 3D video presentations. We collected psycho-physiological evaluations from the participants. The results demonstrated that a wider FOV enhanced the perception of sublimity and pleasantness. In particular, gaze fixation time tended to increase under conditions of wider FOV and 3D presentation, supporting the effect of sublimity in VR. This suggests the potential of VR as a valuable tool for amplifying the experience of sublimity.