js: 识别队伍角色 (#1149)

* Add files via upload

* Add files via upload

* Add files via upload

* Add files via upload

* Add files via upload

更新了读取五位数的原石和4位数的纠缠

* Add files via upload

减少了空间占用

* Add files via upload

* Delete repo/js/OCR读取当前原石以及纠缠之缘记录并发送通知 directory

* Add files via upload

* Add files via upload

* Delete 识别队伍角色 directory

* Delete repo/js/识别队伍角色 directory

* Add files via upload

* Delete repo/js/识别队伍角色/main.js

* Delete repo/js/识别队伍角色 directory

* Add files via upload

* Add files via upload

* Add files via upload

repo/js/OCR读取主界面队伍/Character_log.txt

@@ -0,0 +1 @@
+

repo/js/OCR读取主界面队伍/main.js

@@ -0,0 +1,246 @@
+// ===== 配置与工具函数 =====
+// 字符替换映射表：修正OCR识别的常见错误字符
+const replacementMap = {
+    "监": "盐",
+    "姐": "妲",
+    "干": "千",
+    "卵": "卯",
+    "凌": "绫",
+    "疑": "凝",
+    "沙": "砂",
+    "谣": "瑶",
+    "萤": "荧",  
+    "霄": "宵"
+};
+
+// 原神角色列表：用于OCR识别后的字符匹配
+const genshinImpactCharacters = [
+    "荧", "空", "神里绫华", "琴", "丽莎", "芭芭拉", "凯亚", "迪卢克", "雷泽", "安柏", "温迪", "香菱",
+    "北斗", "行秋", "魈", "凝光", "可莉", "钟离", "菲谢尔", "班尼特", "达达利亚", "诺艾尔", "七七",
+    "重云", "甘雨", "阿贝多", "迪奥娜", "莫娜", "刻晴", "砂糖", "辛焱", "罗莎莉亚", "胡桃", "枫原万叶",
+    "烟绯", "宵宫", "托马", "优菈", "雷电将军", "早柚", "珊瑚宫心海", "五郎", "九条裟罗", "荒泷一斗",
+    "八重神子", "鹿野院平藏", "夜兰", "绮良良", "埃洛伊", "申鹤", "云堇", "久岐忍", "神里绫人", "柯莱",
+    "多莉", "提纳里", "妮露", "赛诺", "坎蒂丝", "纳西妲", "莱依拉", "流浪者", "珐露珊", "瑶瑶",
+    "艾尔海森", "迪希雅", "米卡", "卡维", "白术", "琳妮特", "林尼", "菲米尼", "莱欧斯利", "那维莱特",
+    "夏洛蒂", "芙宁娜", "夏沃蕾", "娜维娅", "嘉明", "闲云", "千织", "希格雯", "阿蕾奇诺", "赛索斯",
+    "克洛琳德", "艾梅莉埃", "卡齐娜", "基尼奇", "玛拉妮", "希诺宁", "恰斯卡", "欧洛伦", "玛薇卡",
+    "茜特菈莉", "蓝砚", "梦见月瑞希", "伊安珊", "瓦雷莎", "爱可菲", "伊法", "丝柯克", "塔利雅"
+];
+
+// 预加载所有模板图像：避免循环中重复读取文件，提升性能
+const templates = {
+    wanderer: file.ReadImageMatSync("assets/wanderer_icon.png"),        // 流浪者图标（活跃状态）
+    wandererInactive: file.ReadImageMatSync("assets/wanderer_icon_no_active.png"), // 流浪者图标（非活跃状态）
+    yin: file.ReadImageMatSync("assets/yin.png"),                      // 荧的图标模板1
+    yin2: file.ReadImageMatSync("assets/yin2.png"),                    // 荧的图标模板2
+};
+
+// 创建模板匹配对象的工具函数：统一参数格式，减少重复代码
+function createTemplateMatch(templateKey, x, y, width, height) {
+    return RecognitionObject.TemplateMatch(
+        templates[templateKey],  // 使用预加载的模板图像
+        x, y,                    // 识别区域左上角坐标
+        width, height            // 识别区域宽高
+    );
+}
+
+// 过滤非中文字符的工具函数：处理OCR结果中的无效字符
+function filterChineseChars(text) {
+    return text?.replace(/[^\u4e00-\u9fa5]/g, '') || '';  // 保留中文字符，其他字符替换为空
+}
+
+// 图像识别函数：返回布尔值表示是否识别成功
+async function recognizeImage(recognitionObject) {
+    await sleep(500);  // 延迟500ms，避免识别请求过于频繁
+    try {
+        const imageResult = captureGameRegion().find(recognitionObject);
+        // 当识别结果存在且坐标不为(0,0)时（排除无效识别）
+        return !!imageResult && imageResult.x !== 0 && imageResult.y !== 0;
+    } catch (error) {
+        log.error(`识别图像时发生异常: ${error.message}`);
+        return false;
+    }
+}
+
+// OCR文本识别函数：在指定区域重试识别直到超时
+async function recognizeTextInRegion(ocrRegion, timeout = 5000) {
+    let startTime = Date.now();  // 记录开始时间
+    let retryCount = 0;          // 重试次数计数
+
+    // 循环直到超时
+    while (Date.now() - startTime < timeout) {
+        try {
+            // 创建OCR识别区域对象
+            const region = RecognitionObject.ocr(
+                ocrRegion.x, ocrRegion.y, ocrRegion.width, ocrRegion.height
+            );
+            // 绘制识别区域红框（调试用）
+            //captureGameRegion().find(region).DrawSelf("debug");
+
+            const ocrResult = captureGameRegion().find(region);
+            if (ocrResult) {
+                // 应用字符替换映射表修正识别结果
+                let correctedText = ocrResult.text;
+                for (const [wrongChar, correctChar] of Object.entries(replacementMap)) {
+                    correctedText = correctedText.replace(new RegExp(wrongChar, 'g'), correctChar);
+                }
+                return correctedText;  // 返回修正后的文本
+            } else {
+                log.warn(`OCR 识别区域未找到内容`);
+                return null;
+            }
+        } catch (error) {
+            // 记录重试日志
+            retryCount++;
+            log.warn(`OCR 识别失败，正在进行第 ${retryCount} 次重试...`);
+        }
+        await sleep(500);  // 每次重试间隔500ms
+    }
+    log.warn(`经过多次尝试，仍然无法在指定区域识别到文字`);
+    return null;
+}
+
+// 定义识别对象
+const paimonMenuRo = RecognitionObject.TemplateMatch(
+    file.ReadImageMatSync("assets/paimon_menu.png"),
+    0,
+    0,
+    genshin.width / 3.0,
+    genshin.width / 5.0
+);
+
+// 判断是否在主界面的函数
+const isInMainUI = () => {
+    let captureRegion = captureGameRegion();
+    let res = captureRegion.Find(paimonMenuRo);
+    return !res.isEmpty();
+};
+
+// ===== 主流程逻辑 =====
+(async function () {
+    // 初始化游戏界面（设置分辨率等）
+    setGameMetrics(1920, 1080, 1);
+    await genshin.returnMainUi();  // 返回游戏主界面
+    await sleep(2000);  // 等待界面响应
+
+    if (isInMainUI(paimonMenuRo)) {
+
+
+        // 按下ESC键（可能用于打开菜单或其他操作）
+        keyPress("ESCAPE");
+        await sleep(1500);  // 等待界面响应
+
+        // 识别玩家名称（主角名称）
+        const playerNameRegion = { x: 270, y: 40, width: 400, height: 40 };
+        let playerName = await recognizeTextInRegion(playerNameRegion);
+        playerName = filterChineseChars(playerName);  // 过滤非中文字符
+        log.info(`识别到的主角为：${playerName}`);
+
+        // 返回主界面并等待加载
+        await genshin.returnMainUi();
+        await sleep(1000);
+
+        if (!isInMainUI(paimonMenuRo)) {
+            log.error(`返回主界面失败`);
+            return;
+        }
+
+        // 初始化角色识别结果数组
+        const CharacterJudgment = [false, false, false, false];  // 记录每个角色是否识别成功
+        const Character = [];                                    // 存储识别到的角色名称
+
+        // 并行识别4个角色（使用Promise.all提升效率）
+        const recognitionPromises = Array(4).fill().map(async (_, i) => {
+            try {
+                // 计算第i个角色的OCR识别区域（垂直排列，间隔100px）
+                const ocrRegion = {
+                    x: 1620,
+                    y: 225 + i * 100,
+                    width: 160,
+                    height: 60
+                };
+                // 执行OCR识别并过滤字符
+                const recognizedText = await recognizeTextInRegion(ocrRegion);
+                const filteredText = filterChineseChars(recognizedText);
+                //log.info(`[角色${i + 1}] 识别到的文本：${filteredText}`);
+
+                // 与角色列表进行匹配
+                let isCharacterMatched = false;
+                for (let j = 0; j < genshinImpactCharacters.length; j++) {
+                    if (genshinImpactCharacters[j] === filteredText) {
+                        // 匹配成功，记录结果
+                        CharacterJudgment[i] = true;
+                        log.info(`[角色${i + 1}] ${filteredText} 匹配成功`);
+                        Character[i] = filteredText;
+                        await sleep(500);  // 等待后续操作
+                        isCharacterMatched = true;
+                        break;
+                    }
+                }
+
+                // 当OCR未匹配到角色时，尝试通过图像模板识别（散兵/旅行者）
+                if (!isCharacterMatched) {
+                    const { x, y, width, height } = ocrRegion;  // 解构区域坐标
+
+                    // 创建4个模板匹配对象（流浪者和旅行者的不同状态）
+                    const wanderer1 = createTemplateMatch("wanderer", x + 100, y, width + 20, height + 20);
+                    const wanderer2 = createTemplateMatch("wandererInactive", x + 100, y - 10, width + 20, height + 30);
+                    const yin3 = createTemplateMatch("yin", x + 100, y - 10, width + 20, height + 30);
+                    const yin4 = createTemplateMatch("yin2", x + 100, y - 10, width + 20, height + 30);
+
+                    // 执行图像识别并处理结果（添加catch避免异常中断）
+                    const result1 = await recognizeImage(wanderer1).catch(() => false);
+                    const result2 = await recognizeImage(wanderer2).catch(() => false);
+
+                    if (result1 || result2) {
+                        // 识别到流浪者
+                        Character[i] = "流浪者";
+                        log.info(`[角色${i + 1}] 流浪者匹配成功`);
+                        CharacterJudgment[i] = true;
+                    } else {
+                        // 尝试识别旅行者（荧或空）
+                        const result3 = await recognizeImage(yin3).catch(() => false);
+                        const result4 = await recognizeImage(yin4).catch(() => false);
+                        if ((result3 || result4) && filteredText === playerName) {
+                            Character[i] = "荧";
+                            log.info(`[角色${i + 1}] 荧匹配成功`);
+                            CharacterJudgment[i] = true;
+                        } else if (filteredText === playerName) {
+                            Character[i] = "空";
+                            log.info(`[角色${i + 1}] 空匹配成功`);
+                            CharacterJudgment[i] = true;
+                        } else {
+                            log.error(`[角色${i + 1}] 散兵/旅行者识别失败`);
+                        }
+                    }
+                }
+            } catch (error) {
+                // 记录详细错误到日志文件
+                const errorMsg = `[角色${i + 1}] 识别失败: ${error.message}`;
+                file.WriteTextSync("Resources_log.txt", errorMsg + "\n", true);
+                log.warn(errorMsg);
+            }
+        });
+
+        // 等待所有角色识别完成
+        await Promise.all(recognitionPromises);
+
+        // 检查是否全部识别成功
+        const allRecognized = CharacterJudgment.every(Boolean);
+        if (allRecognized) {
+            // 记录识别结果到日志文件
+            const now = new Date();
+            const logContent = `${now.toLocaleString()} —— 1: ${Character[0]} ——- 2:${Character[1]} ——- 3:${Character[2]} ———— 4:${Character[3]}\n`;
+            const writeSuccess = file.WriteTextSync("Character_log.txt", logContent, true);
+            log.info(writeSuccess ? "成功将队伍信息写入日志文件" : "写入日志文件失败");
+        } else {
+            log.error("未能完全识别队伍角色");
+        }
+
+        // 等待后返回主界面
+        await sleep(500);
+        await genshin.returnMainUi();
+    } else {
+        log.error(`返回主界面失败,跳过识别`);
+    }
+})();

repo/js/OCR读取主界面队伍/manifest.json

@@ -0,0 +1,12 @@
+{
+  "manifest_version": 1,
+  "name": "OCR主界面读取队伍 ",
+  "version": "1.0",
+  "description": "至少0.44.3版本。OCR识别队伍角色",
+  "authors": [
+    {
+      "name": "未知_"
+    }
+  ],
+  "main": "main.js"
+}

repo/js/OCR读取当前抽卡资源并发送通知/Resources_log.txt

@@ -0,0 +1 @@
+

repo/js/OCR读取当前抽卡资源并发送通知/main.js

@@ -113,10 +113,10 @@ async function recognizeTextInRegion(ocrRegion, timeout = 5000) {
     // 如果识别到了“常驻祈愿”图标，则识别“原石以及纠缠之缘到数值”
     if (recognized) {
         //原石
-        let ocrRegionYuanShi = { x: 1480, y: 30, width: 160, height: 36 }; // 设置对应的识别区域
+        let ocrRegionYuanShi = { x: 1470, y: 25, width: 180, height: 46 }; // 设置对应的识别区域
         let recognizedText1 = await recognizeTextInRegion(ocrRegionYuanShi);
         //纠缠之缘
-        let ocrRegionInterwinedFate = { x: 1660, y: 30, width: 120, height: 36 }; // 设置对应的识别区域
+        let ocrRegionInterwinedFate = { x: 1650, y: 25, width: 140, height: 46 }; // 设置对应的识别区域
         let recognizedText2 = await recognizeTextInRegion(ocrRegionInterwinedFate);
 
 

repo/js/OCR读取当前抽卡资源并发送通知/manifest.json

@@ -1,7 +1,7 @@
 {
   "manifest_version": 1,
   "name": "读取当前原石以及纠缠之缘记录并发送通知",
-  "version": "1.2",
+  "version": "1.4",
   "description": "至少0.44.3版本。OCR识别原石以及纠缠之缘数值输出到本地",
   "authors": [
     {