\u901a\u8fc7\u5728\u8fb9\u7f18\u4fa7\u5b9e\u73b0\u9ad8\u5ea6\u590d\u6742\u7684\u5b9e\u65f6\u89c6\u9891\u6d41\u5904\u7406\u548c\u51b3\u7b56\uff0c\u5d4c\u5165\u5f0f\u89c6\u89c9\u4eba\u5de5\u667a\u80fd\u5e94\u7528\u6b63\u5728\u4e3a\u5404\u884c\u5404\u4e1a\u5e26\u6765\u53d8\u9769\u3002\u5e94\u7528\u8303\u56f4\u4ece\u81ea\u52a8\u9a7e\u9a76\u5230\u667a\u80fd\u5236\u9020\uff0c\u5728\u8fd9\u4e9b\u5e94\u7528\u4e2d\uff0c\u5feb\u901f\u5206\u6790\u89c6\u89c9\u6570\u636e\u81f3\u5173\u91cd\u8981\u3002\u8fd9\u4e9b\u5e94\u7528\u8981\u6c42\u5b9e\u65f6\u5904\u7406\u3001\u4f4e\u5ef6\u8fdf\u548c\u4f4e\u529f\u8017\uff0c\u4ee5\u6709\u6548\u5206\u6790\u548c\u89e3\u91ca\u89c6\u89c9\u6570\u636e\u3002<\/p>\n\n\n\n Selecting the right hardware platform is crucial in meeting the tight requirements of embedded AI applications. The requirements involve achieving high-performance video processing with minimal and deterministic latency while maintaining low power consumption. Another requirement of the AI at-the-edge applications is the interconnectivity with multiple sensors and other peripheral devices. Traditional processors, such as CPUs and GPUs, often struggle to meet these requirements; CPUs, designed for sequential processing, can not provide the processing bandwidth of a parallelized processor architecture; GPUs, even though based on highly parallelized architecture, struggle to achieve power efficiency and ultra-low, deterministic latency; neither technology offers sophisticated sensor-fusion capabilities. ASICs, although able to achieve ultra-low latency, high power efficiency and sensor fusion capability, are not suited for applications that require flexibility to adapt to constantly evolving AI algorithms.<\/p>\n\n\n\n FPGA technology offers a convincing solution to the challenges of the embedded vision AI:<\/p>\n\n\n\n To demonstrate the capabilities of embedded AI on FPGA technology, the Enclustra Design Services team implemented a vision AI application on an Enclustra FPGA\/SoC module, Mercury+ XU7<\/a>, and Enclustra Mercury+ ST1<\/a> base board. Enclustra team combined two available AI models, image classification (YoloV3-based) and car colour detection in a sequence, to achieve maximal algorithm performance. The models were implemented using the AMD Vitis\u2122 AI toolchain, which translates AI models for optimal implementation in FPGA\/SoC fabric and enables optimal usage of SoC resources.<\/p>\n\n\n\n In summary, FPGA technology\u2019s unique mix of performance, efficiency, and adaptability positions them as the preferred hardware choice for next-generation embedded AI applications. Enclustra team\u2019s deep experience in FPGA technology and FPGA development tools transforms the initial idea into reality.<\/p>\n\n\n\n Vision AI application on an Enclustra FPGA\/SoC module and base board combination (Mercury+ XU7 and Mercury+ ST1)<\/em><\/p>\n\n\n\n Two AI models in a sequence: YoloV3-based image classification and car color detection<\/em><\/p>","protected":false},"excerpt":{"rendered":" \u901a\u8fc7\u5728\u8fb9\u7f18\u4fa7\u5b9e\u73b0\u9ad8\u5ea6\u590d\u6742\u7684\u5b9e\u65f6\u89c6\u9891\u6d41\u5904\u7406\u548c\u51b3\u7b56\uff0c\u5d4c\u5165\u5f0f\u89c6\u89c9\u4eba\u5de5\u667a\u80fd\u5e94\u7528\u6b63\u5728\u4e3a\u5404\u884c\u5404\u4e1a\u5e26\u6765\u53d8\u9769\u3002\u5e94\u7528\u8303\u56f4\u4ece\u81ea\u52a8\u9a7e\u9a76\u5230\u667a\u80fd\u5236\u9020\uff0c\u5728\u8fd9\u4e9b\u5e94\u7528\u4e2d\uff0c\u5feb\u901f\u5206\u6790\u89c6\u89c9\u6570\u636e\u81f3\u5173\u91cd\u8981\u3002\u8fd9\u4e9b\u5e94\u7528\u8981\u6c42\u5b9e\u65f6\u5904\u7406\u3001\u4f4e\u5ef6\u8fdf\u548c\u4f4e\u529f\u8017\uff0c\u4ee5\u6709\u6548\u5206\u6790\u548c\u89e3\u91ca\u89c6\u89c9\u6570\u636e\u3002<\/p>","protected":false},"featured_media":2697,"template":"","meta":{"_acf_changed":true,"_seopress_robots_primary_cat":"","_seopress_titles_title":"Revolutionizing Vision AI Applications with FPGA Technology","_seopress_titles_desc":"Embedded vision AI applications are revolutionizing industries by enabling highly complex real-time video stream processing and decision-making at the edge.","_seopress_robots_index":"yes"},"story-category":[80],"class_list":["post-2693","success-story","type-success-story","status-publish","has-post-thumbnail","hentry","story-category-ai-machine-learning"],"acf":[],"_links":{"self":[{"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/success-story\/2693","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/success-story"}],"about":[{"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/types\/success-story"}],"version-history":[{"count":0,"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/success-story\/2693\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/media\/2697"}],"wp:attachment":[{"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/media?parent=2693"}],"wp:term":[{"taxonomy":"story-category","embeddable":true,"href":"https:\/\/test.enclustra.com.cn\/zh\/wp-json\/wp\/v2\/story-category?post=2693"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"article](\"https:\/\/test.enclustra.com.cn\/wp-content\/uploads\/2025\/04\/article_vision_ai_website-750x-669-1.jpg\" "\\"1.jpg\\"")
<\/figure>\n\n\n\n\u6311\u6218<\/strong><\/h2>\n\n\n\n
\u89e3\u51b3\u65b9\u6848<\/strong><\/h2>\n\n\n\n
\n
Example Application<\/strong><\/h2>\n\n\n\n
![\"ai](\"https:\/\/test.enclustra.com.cn\/wp-content\/uploads\/2025\/04\/ai-demo_diagram_2-1800x-a24.jpg\" "\\"5.jpg\\"")
<\/figure>\n\n\n\n![\"ai](\"https:\/\/test.enclustra.com.cn\/wp-content\/uploads\/2025\/04\/ai-demo_diagram_1-1800x-155.jpg\")
<\/figure>\n\n\n\n