This document introduces a deep learning-based image segmentation method. The method uses a convolutional neural network (CNN) to segment images. A CNN is a deep learning model that can learn the features of images in order to segment them. The method first divides the image into multiple small regions, then uses the CNN to segment each region. Finally, the segmentation results of all regions are merged together to obtain the segmentation result of the entire image.
The JE12 latching relay is designed for high‑power applications, offering a 120 A switching capability, heavy load capacity up to 33.24 kVA, and a dielectric strength of 4 kV between coil and contacts. It features a compact size of 52 × 43 × 22 mm and a weight of about 100 g, making it suitable for space‑constrained installations.
Ordering options include single‑coil (2.4 W) or double‑coil (4.8 W) versions, multiple contact terminal types (A‑E), and various coil voltages (6, 12, 24, 48 VDC). The relay complies with RoHS and IEC 62055‑31 standards, and operates reliably within an ambient temperature range of –40 °C to 70 °C.
| Switching capability | 120 A |
|---|---|
| Dielectric strength | 4 kV |
| Heavy load rating | 33.24 kVA |
| Single‑coil power | 2.4 W |
| Double‑coil power | 4.8 W |
| Contact material | Silver alloy |
| Insulation resistance | 1000 MΩ |
| Ambient temperature range | -40 °C to 70 °C |
| Humidity tolerance | 98 %RH at 40 °C |
| Dimensions | 52.0 × 43.0 × 22.0 mm |
| Unit weight | Approx. 100 g |
| Maximum switching voltage | 440 VAC |
| Maximum switching power | 33 240 VA / 3 360 W |
| Operate time | 20 ms max |
| Release time | 20 ms max |
The JE12 relay is rated for a switching current of 120 A.
It provides a minimum dielectric strength of 4 kV (4000 VAC) between the coil and contacts.
The relay can operate from –40 °C up to 70 °C.