优化图像分类：实现“未检测到”结果的策略与实践

private void classifyImage(Bitmap image) {
    try {
        FruitDisease model = FruitDisease.newInstance(getApplicationContext());

        // ... (省略图片预处理和ByteBuffer填充部分，与原代码相同)
        TensorBuffer inputFeature = TensorBuffer.createFixedSize(new int[]{1, 224, 224, 3}, DataType.FLOAT32);
        ByteBuffer byteBuffer = ByteBuffer.allocateDirect(4 * imageSize * imageSize * 3);
        byteBuffer.order(ByteOrder.nativeOrder());

        int[] intValue = new int[imageSize * imageSize];
        image.getPixels(intValue, 0, image.getWidth(), 0, 0, image.getWidth(), image.getHeight());

        int pixel = 0;
        for (int i = 0; i < imageSize; i++) {
            for (int j = 0; j < imageSize; j++) {
                int val = intValue[pixel++];
                // 假设模型输入要求归一化到0-1
                byteBuffer.putFloat(((val >> 16) & 0xFF) * (1.f / 255.f)); // R
                byteBuffer.putFloat(((val >> 8) & 0xFF) * (1.f / 255.f));  // G
                byteBuffer.putFloat((val & 0xFF) * (1.f / 255.f));        // B
            }
        }
        inputFeature.loadBuffer(byteBuffer);

        // 执行推理
        FruitDisease.Outputs outputs = model.process(inputFeature);
        TensorBuffer outputFeature0 = outputs.getOutputFeature0AsTensorBuffer();

        float[] confidences = outputFeature0.getFloatArray(); // 获取所有类别的置信度

        // 找到置信度最高的类别及其置信度值
        int maxPos = 0;
        float maxConfidence = 0;
        for (int i = 0; i < confidences.length; i++) {
            if (confidences[i] > maxConfidence) {
                maxConfidence = confidences[i];
                maxPos = i;
            }
        }

        // 定义类别标签
        String[] classes = {"Watermelon Healthy", "Watermelon Blossom End Rot", "Watermelon Anthracnose",
                "Mango Healthy", "Mango Bacterial Canker", "Mango Anthracnose",
                "Orange Scab", "Orange Healthy",
                "Orange Bacterial Citrus Canker", "Banana Healthy", "Banana Crown Rot",
                "Banana Anthracnose", "Apple Scab", "Apple Healthy", "Apple Black Rot Canker"};

        // --- 核心逻辑：基于置信度阈值判断是否为“未检测到” ---
        // 设置一个置信度阈值，例如0.7 (70%)。这个值需要根据模型的实际表现进行调整和优化。
        float confidenceThreshold = 0.7f; // 阈值范围通常在0到1之间

        if (maxConfidence < confidenceThreshold) {
            // 如果最高置信度低于阈值，则认为未检测到目标对象
            result.setText("未检测到目标对象 (No target object detected)");
        } else {
            // 否则，显示检测到的类别
            result.setText(classes[maxPos]);
        }
        // --- 结束核心逻辑 ---

        // 构建并显示所有类别的详细置信度
        StringBuilder confidenceDetails = new StringBuilder();
        for (int i = 0; i < classes.length; i++) {
            confidenceDetails.append(String.format("%s: %.1f%%\n", classes[i], confidences[i] * 100));
        }
        confidence.setText(confidenceDetails.toString());
        confidence.setVisibility(View.VISIBLE);

        // 释放模型资源
        model.close();

    } catch (IOException e) {
        // TODO Handle the exception
        e.printStackTrace();
    }
}